DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claim 10 is rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
As to claim 10
The limitation “Design knowledge needed to answer the answer the natural language query” makes the claim unclear on what answer the answer the answer the natural language represents. Examiner suggests the limitation should read as “ Design knowledge needed to answer the natural language query.”
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-20 are rejected under 35 USC § 101 because claimed invention is directed to the abstract idea without significantly more.
As to claim 1:
Step 1 Analysis: Is the claim to a process, machine, manufacture or composition of matter? See MPEP § 2106.03.
Claim 1 is a method claim, therefore it falls under one of four categories of statutory subject matter.
Step 2A Prong One Analysis: Does the claim recite an abstract idea, law of nature, or natural phenomenon? See MPEP § 2106.04(II)(A)(1).
The limitation “building a prompt for a large language model (LLM) using the natural language query and the retrieved portion of the collected knowledge. ” is an abstract idea of a mental process that can practically be performed in the human mind, with or without the use of physical aid such as pen and paper (including an observation, evaluation, judgment, opinion). See MPEP § 2106.04(a)(2)(III).
Step 2A Prong Two Analysis: Does the claim recite additional elements that integrate the judicial exception into a practical application? See MPEP § 2106.04(d).
The limitation “A method performed by one or more computer systems for implementing a virtual solution architect (VSA), the method comprising: ” is an additional element that amounts to adding the words “apply it” (or an equivalent) with the judicial exception or merely uses a computer in its ordinary capacity as a tool to perform an existing process. See MPEP §§ 2106.04(d), 2106.05(f)(2).
The limitation “collecting knowledge relevant to technology solution architecture design, the collected knowledge including” is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering to the judicial exception. The claim recites gathering information from variety of sources See MPEP §§ 2106.04(d), 2106.05(g).
The limitation “a knowledge graph comprising business processes and relationships between each business process and other entities.” is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering to the judicial exception. The claim recites gathering information from variety of sources. See MPEP §§ 2106.04(d), 2106.05(g).
The limitation “a first set of text chunks and associated embeddings, the first set of text chunks originating from application programming interface (API) documentation for a plurality of packaged business capabilities (PBCs).” is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering to the judicial exception. The claim recites gathering information from API. See MPEP §§ 2106.04(d), 2106.05(g).
The limitation “and a second set of text chunks and associated embeddings, the second set of text chunks originating from one or more help websites of one or more vendors of the PBCs. ” is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering to the judicial exception. The claim recites gathering information from internet sites. See MPEP §§ 2106.04(d), 2106.05(g).
The limitation “receiving a natural language query pertaining to a technology solution architecture.” is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering to the judicial exception. The claim recites receiving instructions in the form of natural language. See MPEP §§ 2106.04(d), 2106.05(g).
The limitation “retrieving at least a portion of the collected knowledge based on the natural language query “is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering to the judicial exception. The claim recites extracting a part of collected information based on the query received. See MPEP §§ 2106.04(d), 2106.05(g).
The limitation “ providing prompt as input to the LLM, thereby causing the LLM to output a natural language answer to the natural language query “is an additional element that amounts to adding insignificant extra-solution activity of mere data input /output to the judicial exception. The claim limitation recites providing input to model and generating an output based on input(i.e. natural language ).See MPEP §§ 2106.04(d), 2106.05(g).
The limitation “ and providing the natural language answer to an originator of the natural language query. “Is an additional element that amounts to adding insignificant extra-solution activity of mere data input /output to the judicial exception. The claim limitation recites providing the output in the form of answer to the user. See MPEP §§ 2106.04(d)
Step 2B Analysis: Does the claim recite additional elements that amount to significantly more than the judicial exception? See MPEP § 2106.05.
The limitation “A method performed by one or more computer systems for implementing a virtual solution architect (VSA), the method comprising: ” is an additional element that amounts to adding the words “apply it” (or an equivalent) with the judicial exception or merely uses a computer in its ordinary capacity as a tool to perform an existing process. Furthermore, the additional element is directed to a method Performed by computational unit, which the courts have recognized as well‐understood, routine, and conventional when they are claimed in a generic manner. See MPEP § 2106.05(d)(II).
The limitation “collecting knowledge relevant to technology solution architecture design, the collected knowledge including” is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering to the judicial exception. The claim recites gathering information from variety of sources which the courts have recognized as well‐understood, routine, and conventional when they are claimed in a generic manner. See MPEP § 2106.05(d)(II).
The limitation “a knowledge graph comprising business processes and relationships between each business process and other entities.” is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering to the judicial exception. The claim recites gathering information from variety of sources which the courts have recognized as well‐understood, routine, and conventional when they are claimed in a generic manner. See MPEP § 2106.05(d)(II).
The limitation “a first set of text chunks and associated embeddings, the first set of text chunks originating from application programming interface (API) documentation for a plurality of packaged business capabilities (PBCs).” is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering to the judicial exception. The claim recites gathering information from API which the courts have recognized as well‐understood, routine, and conventional when they are claimed in a generic manner. See MPEP § 2106.05(d)(II).
The limitation “and a second set of text chunks and associated embeddings, the second set of text chunks originating from one or more help websites of one or more vendors of the PBCs. ” is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering to the judicial exception. The claim recites gathering information from internet sites. which the courts have recognized as well‐understood, routine, and conventional when they are claimed in a generic manner. See MPEP § 2106.05(d)(II).
The limitation “receiving a natural language query pertaining to a technology solution architecture.” is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering to the judicial exception. The claim recites receiving instructions in the form of natural language which the courts have recognized as well‐understood, routine, and conventional when they are claimed in a generic manner. See MPEP §§ 2106.04(d), 2106.05(g).
The limitation “retrieving at least a portion of the collected knowledge based on the natural language query “is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering to the judicial exception. The claim recites extracting a part of collected information based on the query received which the courts have recognized as well‐understood, routine, and conventional when they are claimed in a generic manner. See MPEP § 2106.05(d)(II).
The limitation “ providing prompt as input to the LLM, thereby causing the LLM to output a natural language answer to the natural language query “is an additional element that amounts to adding insignificant extra-solution activity of mere data input /output to the judicial exception. The claim limitation recites providing input to model and generating an output based on input(i.e. natural language ) which the courts have recognized as well‐understood, routine, and conventional when they are claimed in a generic manner. See MPEP § 2106.05(d)(II). See MPEP §§ 2106.04(d), 2106.05(g).
The limitation “ and providing the natural language answer to an originator of the natural language query. “Is an additional element that amounts to adding insignificant extra-solution activity of mere data input /output to the judicial exception. The claim limitation recites providing the output in the form of answer to the user which the courts have recognized as well‐understood, routine, and conventional when they are claimed in a generic manner. See MPEP § 2106.05(d)(II).
Therefore, in examining elements as recited by the limitations individually and as an ordered combination, as a whole the independent claim limitations do not recite what the courts have identified as “significantly more”.
As to claim 2
Step 2A Prong Two Analysis: Does the claim recite additional elements that integrate the judicial exception into a practical application? See MPEP § 2106.04(d).
The limitation “ wherein the natural language answer comprises a proposed design for the technology solution architecture. “Is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering. See MPEP §§ 2106.04(d), 2106.05(g).
Step 2B Analysis: Does the claim recite additional elements that amount to significantly more than the judicial exception? See MPEP § 2106.05
The limitation “ wherein the natural language answer comprises a proposed design for the technology solution architecture. “Is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering. which the courts have recognized as well‐understood, routine, and conventional when they are claimed in a generic manner. See MPEP § 2106.05(d)(II).
As to Claim 3
Step 2A Prong Two Analysis: Does the claim recite additional elements that integrate the judicial exception into a practical application? See MPEP § 2106.04(d).
The limitation “wherein the natural language query includes one or more business objectives that should be achieved by the technology solution architecture,” Is an additional element that amounts to adding insignificant extra-solution activity of mere data input/output. See MPEP §§ 2106.04(d), 2106.05(g).
The limitation “,and wherein the proposed design includes a subset of the plurality of PBCs and a plan for integrating the subset in order to meet the one or more business objectives “Is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering. See MPEP §§ 2106.04(d), 2106.05(g).
Step 2B Analysis: Does the claim recite additional elements that amount to significantly more than the judicial exception? See MPEP § 2106.05
The limitation “wherein the natural language query includes one or more business objectives that should be achieved by the technology solution architecture,” Is an additional element that amounts to adding insignificant extra-solution activity of mere data input/output. The Claim limitation recites, the objectives that is to be met by architecture which the courts have recognized as well‐understood, routine, and conventional when they are claimed in a generic manner. See MPEP § 2106.05(d)
The limitation “,and wherein the proposed design includes a subset of the plurality of PBCs and a plan for integrating the subset in order to meet the one or more business objectives “Is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering which the courts have recognized as well‐understood, routine, and conventional when they are claimed in a generic manner. See MPEP § 2106.05(d)(II).
As to claim 4:
Step 2A Prong Two Analysis: Does the claim recite additional elements that integrate the judicial exception into a practical application? See MPEP § 2106.04(d).
The limitation “wherein the other entities include one or more PBCs associated with the business process, one or more business domains to which the business process belongs, and one or more industries that the business process is relevant for” is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering. Furthermore, the claim limitation recites extracting a data/information from Variety of sources. See MPEP §§, 2106.05(g).
Step 2B Analysis: Does the claim recite additional elements that amount to significantly more than the judicial exception? See MPEP § 2106.05.
The limitation “wherein the other entities include one or more PBCs associated with the business process, one or more business domains to which the business process belongs, and one or more industries that the business process is relevant for” is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering. Furthermore, the claim limitation recites including data/information from Variety of sources which the courts have recognized as well‐understood, routine, and conventional when they are claimed in a generic manner. See MPEP §§ 2106.05(g).
See MPEP §§, 2106.05(g).
As to Claim 5:
Step 2A Prong Two Analysis: Does the claim recite additional elements that integrate the judicial exception into a practical application? See MPEP § 2106.04(d).
The limitation “Wherein the collected knowledge further includes a third set of text chunks and associated embeddings, the third set of text chunks originating from one or more electronic documents” Is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering. See MPEP §§ 2106.04(d), 2106.05(g).
Step 2B Analysis: Does the claim recite additional elements that amount to significantly more than the judicial exception? See MPEP § 2106.05
The limitation ““Wherein the collected knowledge further includes a third set of text chunks and associated embeddings, the third set of text chunks originating from one or more electronic documents” Is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering which the courts have recognized as well‐understood, routine, and conventional when they are claimed in a generic manner. See MPEP § 2106.05(d)(II).
As to Claim 6:
Step 2A Prong One Analysis: Does the claim recite an abstract idea, law of nature, or natural phenomenon? See MPEP § 2106.04(II)(A)(1).
The limitation “Converting the non- textual data into further textual data” is an abstract idea of a mental process. Under its broadest reasonable interpretation in light of the specification, (“paragraph [0025] Examples of non-textual data include drawings, diagrams, and tables. Data collector 108 can then convert the non-textual data into a textual format (step 206).”). This limitation encompasses the mental process of converting figures /drawings/ tables into text or document which is an evaluation that is practically capable of being performed in the human mind with the assistance of pen and paper. See MPEP § 2106.04(a)(2)(III).
The limitation “splitting the textual data and the further textual data into a plurality of text chunks; and for each text chunk: generating a database insert statement for inserting content of the text chunk into a knowledge graph database storing the knowledge graph” is an abstract idea of a mental process. (i.e. evaluation) See MPEP § 2106.04(a)(2)(III).
Step 2A Prong Two Analysis: Does the claim recite additional elements that integrate the judicial exception into a practical application? See MPEP § 2106.04(d).
The limitation “wherein collecting the knowledge comprises: extracting textual data from a knowledge source that holds business process relationship information “is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering. when the claim is given its broadest reasonable interpretation in light of the specification ( “paragraph [0024] Sarting with step 202, data collector 108 can extract textual data from knowledge source S (i.e., the source holding business process relationship information). For example, if knowledge source S is a website, data collector 108 can crawl the webpages of the website and extract the text from each webpage. If the knowledge source is a set of documents, VSA backend 106 can extract the text from each document.”). The claim recites gathering data from knowledge sources holding variety of information/data and is considered falling under the grouping of mere data gathering. See MPEP §§, 2106.05(g).
The limitation “extracting non-textual data from the knowledge source. is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering. when the claim is given its broadest reasonable interpretation in light of the specification ( “paragraph [ 0025] At step 204, data collector 108 can extract non-textual data from knowledge source S.”). The claim recites gathering data from knowledge sources holding variety of information/data and is considered falling under the grouping of mere data gathering. See MPEP §§, 2106.05(g).
The limitation “and running the database insert statement on the knowledge graph database.” is an additional element that amounts to adding the words “apply it” (or an equivalent) with the judicial exception or merely uses a computer in its ordinary capacity as a tool to perform an existing process. See MPEP §§ 2106.04(d), 2106.05(f)(2).
Step 2B Analysis: Does the claim recite additional elements that amount to significantly more than the judicial exception? See MPEP § 2106.05.
The limitation “wherein collecting the knowledge comprises: extracting textual data from a knowledge source that holds business process relationship information “is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering. when the claim is given its broadest reasonable interpretation in light of the specification ( “paragraph [0024] Sarting with step 202, data collector 108 can extract textual data from knowledge source S (i.e., the source holding business process relationship information). For example, if knowledge source S is a website, data collector 108 can crawl the webpages of the website and extract the text from each webpage. If the knowledge source is a set of documents, VSA backend 106 can extract the text from each document.”). The claim recites gathering data from knowledge sources holding variety of information/data and is considered falling under the grouping of mere data gathering. Furthermore, the additional elements are recognized by the courts as well‐understood, routine, and conventional when they are claimed in a generic manner. See MPEP § 2106.05(d).
The limitation “extracting non-textual data from the knowledge source. is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering. when the claim is given its broadest reasonable interpretation in light of the specification ( “paragraph [ 0025] At step 204, data collector 108 can extract non-textual data from knowledge source S.”). The claim recites gathering data from knowledge sources holding variety of information/data and is considered falling under the grouping of mere data gathering. . Furthermore, the additional elements are recognized by the courts as well‐understood, routine, and conventional when they are claimed in a generic manner See MPEP § 2106.05(d).
The limitation “and running the database insert statement on the knowledge graph database.” is an additional element that amounts to adding the words “apply it” (or an equivalent) with the judicial exception or merely uses a computer in its ordinary capacity as a tool to perform an existing process which the courts have recognized as well‐understood, routine, and conventional when they are claimed in a generic manner. See MPEP § 2106.05(d)(II).
As to Claim 7:
Step 2A Prong One Analysis: Does the claim recite an abstract idea, law of nature, or natural phenomenon? See MPEP § 2106.04(II)(A)(1).
The limitation “wherein the database insert statement is generated by providing the content of the text chunk and a schema of the knowledge graph database to the LLM.” is an abstract idea of a mental process. (i.e. evaluation) practically be performed in the human mind, with or without the use of a physical aid such as pen and paper See MPEP § 2106.04(a)(2)(III).
As to Claim 8:
Step 2A Prong One Analysis: Does the claim recite an abstract idea, law of nature, or natural phenomenon? See MPEP § 2106.04(II)(A)(1).
The limitation “converting the non-textual data into further textual data; splitting the textual data and the further textual data into a plurality of text chunks; and for each text chunk:” is an abstract idea of a mental process. (i.e. evaluation) practically be performed in the human mind, with or without the use of physical aid such as pen and paper. See MPEP § 2106.04(a)(2)
The limitation “and for each text chunk: creating an embedding of the text chunk” is the abstract idea of a mathematical calculation, as directed to “a claim that recites a mathematical calculation, when the claim is given its broadest reasonable interpretation in light of the specification, (“paragraph ,[0034] data collector 108 can create an embedding of the text chunk, where the embedding is a dense vector that represents the semantic content of the text chunk in a mathematical format (step 412).”) will be considered as falling within the "mathematical concepts grouping. See MPEP § 2106.04(a)(2)(I)(C).
Step 2A Prong Two Analysis: Does the claim recite additional elements that integrate the judicial exception into a practical application? See MPEP § 2106.04(d).
The limitation “wherein collecting the knowledge comprises: extracting textual data from the API documentation and the one or more help websites extracting non-textual data from the API documentation and the one or more help websites” is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering. Furthermore, the claim limitation recites extracting a data/information from Variety of sources. See MPEP §§, 2106.05(g).
The limitation “ and storing the text chunk and the embedding in a vector database of the VSA” is an additional element that amounts to adding insignificant extra-solution activity of mere data storing. See MPEP §§ 2106.05(g).
Step 2B Analysis: Does the claim recite additional elements that amount to significantly more than the judicial exception? See MPEP § 2106.05.
The limitation “wherein collecting the knowledge comprises: extracting textual data from the API documentation and the one or more help websites; extracting non-textual data from the API documentation and the one or more help websites” is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering. Furthermore, the claim limitation recites extracting a data/information from Variety of sources. which the courts have recognized as well‐understood, routine, and conventional when they are claimed in a generic manner. See MPEP §§ 2106.05(g).
The limitation “ and storing the text chunk and the embedding in a vector database of the VSA” is an additional element that amounts to adding insignificant extra-solution activity of mere data storing (i.e. storing the associated data into the database). Furthermore, Receiving, transmitting and communicating data are insignificant extra-solution activities that are well-understood, routine, and conventional. See MPEP2106.05(d)(II) (“The courts have recognized the following computer functions as well‐understood, routine, and conventional functions… i. Receiving or transmitting data over a network…iv. Storing and retrieving information in memory”) (citing OIP Techs., Inc., v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1093 (Fed. Cir. 2015)).
As to Claim 9:
Step 2A Prong One Analysis: Does the claim recite an abstract idea, law of nature, or natural phenomenon? See MPEP § 2106.04(II)(A)(1).
The limitation “wherein the embedding is created” is the abstract idea of a mathematical calculation, as directed to “a claim that recites a mathematical calculation, when the claim is given its broadest reasonable interpretation in light of the specification, (“paragraph ,[0034] data collector 108 can create an embedding of the text chunk, where the embedding is a dense vector that represents the semantic content of the text chunk in a mathematical format (step 412).”) will be considered as falling within the "mathematical concepts grouping. See MPEP § 2106.04(a)(2)(I)(C
Step 2A Prong Two Analysis: Does the claim recite additional elements that integrate the judicial exception into a practical application? See MPEP § 2106.04(d).
The limitation “by providing the text chunk as input to an embedding model” is an additional element that amounts to adding insignificant extra-solution activity of data gathering/ forwarding. See MPEP2106.05(d)(II)
Step 2B Analysis: Does the claim recite additional elements that amount to significantly more than the judicial exception? See MPEP § 2106.05.
The limitation “by providing the text chunk as input to an embedding model” is an additional element that amounts to adding insignificant extra-solution activity of data gathering/ forwarding. Furthermore, Receiving, transmitting and communicating data are insignificant extra-solution activities that are well-understood, routine, and conventional. See MPEP2106.05(d)(II) (“The courts have recognized the following computer functions as well‐understood, routine, and conventional functions… i. Receiving or transmitting data over a network…iv. Storing and retrieving information in memory”) (citing OIP Techs., Inc., v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1093 (Fed. Cir. 2015)). Therefore, recitation of “providing the text chunk as input to an embedding mode” is the well-understood, routine, conventional activity of receiving or transmitting data over a network, as discussed in MPEP § 2106.05(d).
As to Claim 10:
Step 2A Prong One Analysis: Does the claim recite an abstract idea, law of nature, or natural phenomenon? See MPEP § 2106.04(II)(A)(1).
The limitation “wherein the retrieving comprises: determining a plan for answering the natural language query; and identifying, from the plan, one or more types of solution design knowledge needed to answer the answer the natural language query.” ” is the abstract idea of a mental process that can practically be performed in the human mind, with or without the use of a physical aid such as pen and paper (including an observation, evaluation, judgment, opinion). See MPEP § 2106.04(a)(2)(III).
As to Claim 11
Step 2A Prong One Analysis: Does the claim recite an abstract idea, law of nature, or natural phenomenon? See MPEP § 2106.04(II)(A)(1).
The limitation “wherein the retrieving further comprises, upon determining that the one or more types of solution design knowledge include business process relationship information: determining, based on the natural language query, a database query statement for querying the business process relationship information in the knowledge graph” is the abstract idea of a mental process that can practically be performed in the human mind, with or without the use of a physical aid such as pen and paper (including an observation, evaluation, judgment, opinion). See MPEP § 2106.04(a)(2)(III).
Step 2A Prong Two Analysis: Does the claim recite additional elements that integrate the judicial exception into a practical application? See MPEP § 2106.04(d).
The limitation.” running the database query statement on a knowledge graph database storing the knowledge graph ” is an additional element that amounts to adding the words “apply it” (or an equivalent) with the judicial exception or merely uses a computer in its ordinary capacity as a tool to perform an existing process. See MPEP §§ 2106.04(d), 2106.05(f)(2).
The limitation “thereby causing the knowledge graph database to output the business process relationship information; and placing the business process relationship information in a context set for the natural language query. ”Is an additional element that amounts to adding insignificant extra-solution activity of mere data output and data storing. The claim limitation recites outputting the business information after running a query(i.e. data output”) and placing the information in context set (i.e. data storing). See MPEP §§ 2106.05(g).
Step 2B Analysis: Does the claim recite additional elements that amount to significantly more than the judicial exception? See MPEP § 2106.05.
The limitation.” running the database query statement on a knowledge graph database storing the knowledge graph ” is an additional element that amounts to adding the words “apply it” (or an equivalent) with the judicial exception or merely uses a computer in its ordinary capacity as a tool to perform an existing process which the courts have recognized as well‐understood, routine, and conventional when they are claimed in a generic manner. See MPEP § 2106.05(d)(II).
The limitation “thereby causing the knowledge graph database to output the business process relationship information; and placing the business process relationship information in a context set for the natural language query. ”Is an additional element that amounts to adding insignificant extra-solution activity of mere data output and data storing. The claim limitation recites outputting the business information after running a query(i.e. data output”) and placing the information in context set (i.e. data storing) which the courts have recognized as well‐understood, routine, and conventional when they are claimed in a generic manner. See MPEP §§ 2106.05(g).
As to Claim 12:
Step 2A Prong One Analysis: Does the claim recite an abstract idea, law of nature, or natural phenomenon? See MPEP § 2106.04(II)(A)(1).
The limitation “wherein the database query statement is generated” is an abstract idea of a mental process. (i.e. evaluation) practically be performed in the human mind, with or without the use of a physical aid such as pen and paper See MPEP § 2106.04(a)(2)(III).
Step 2A Prong Two Analysis: Does the claim recite additional elements that integrate the judicial exception into a practical application? See MPEP § 2106.04(d).
The limitation. “by providing the natural language query and a schema of the knowledge graph database to the LLM” is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering/ forwarding. See MPEP §§ 2106.05(g)
Step 2B Analysis: Does the claim recite additional elements that amount to significantly more than the judicial exception? See MPEP § 2106.05.
The limitation.” ” by providing the natural language query and a schema of the knowledge graph database to the LLM” is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering/ forwarding. Furthermore, receiving, transmitting and communicating data are insignificant extra-solution activities that are well-understood, routine, and conventional. See MPEP2106.05(d)(II) (“The courts have recognized the following computer functions as well‐understood, routine, and conventional functions… i. Receiving or transmitting data over a network…iv. Storing and retrieving information in memory”) (citing OIP Techs., Inc., v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1093 (Fed. Cir. 2015))
As to Claim 13:
Step 2A Prong One Analysis: Does the claim recite an abstract idea, law of nature, or natural phenomenon? See MPEP § 2106.04(II)(A)(1).
The limitation “wherein the retrieving further comprises, upon determining that the one or more types of solution design knowledge include other information that is not business process relationship information: performing a first similarity match between the embedding of the natural language query and the associated embeddings of the first set of text chunks, the performing of the first similarity match resulting in identification of one or more first text chunks in the first set that are relevant to the natural language query” is the abstract idea of a mental process that can practically be performed in the human mind, with or without the use of a physical aid such as pen and paper (including an observation, evaluation, judgment, opinion). See MPEP § 2106.04(a)(2)(III).
The limitation “creating an embedding of the natural language query” is the abstract idea of a mathematical calculation, as directed to “a claim that recites a mathematical calculation, when the claim is given its broadest reasonable interpretation in light of the specification, (“paragraph ,[0034] data collector 108 can create an embedding of the text chunk, where the embedding is a dense vector that represents the semantic content of the text chunk in a mathematical format (step 412).”) will be considered as falling within the "mathematical concepts grouping. See MPEP § 2106.04(a)(2)(I)(C
Step 2A Prong Two Analysis: Does the claim recite additional elements that integrate the judicial exception into a practical application? See MPEP § 2106.04(d).
The limitation.” and placing the one or more first text chunks in the context set. ” is an additional element that amounts to adding insignificant extra-solution activity of mere data output and data storing. The claim limitation recites placing the data in context set. See MPEP §§ 2106.05(g)
Step 2B Analysis: Does the claim recite additional elements that amount to significantly more than the judicial exception? See MPEP § 2106.05.
The limitation.” and placing the one or more first text chunks in the context set. ” is an additional element that amounts to adding insignificant extra-solution activity of mere data output and data storing. The claim limitation recites placing the data in context set. which the courts have recognized as well‐understood, routine, and conventional when they are claimed in a generic manner. See MPEP §§ 2106.05(g)
As to claim 14
Step 2A Prong One Analysis: Does the claim recite an abstract idea, law of nature, or natural phenomenon? See MPEP § 2106.04(II)(A)(1).
The limitation “wherein the embedding is created by” is the abstract idea of a he abstract idea of a mathematical relationship, as directed to “a mathematical relationship is a relationship between variables or numbers. A mathematical relationship may be expressed in words or using mathematical symbols”. See MPEP § 2106.04(a)(2)(I)(A).
Step 2A Prong Two Analysis: Does the claim recite additional elements that integrate the judicial exception into a practical application? See MPEP § 2106.04(d).
The limitation “by providing the natural language query as input to an embedding model” is an additional element that amounts to adding insignificant extra-solution activity of data gathering/ forwarding. See MPEP2106.05(d)(II)
Step 2B Analysis: Does the claim recite additional elements that amount to significantly more than the judicial exception? See MPEP § 2106.05.
The limitation “by providing the natural language input to an embedding model” is an additional element that amounts to adding insignificant extra-solution activity of data gathering/ forwarding. Furthermore, Receiving, transmitting and communicating data are insignificant extra-solution activities that are well-understood, routine, and conventional. See MPEP2106.05(d)(II) (“The courts have recognized the following computer functions as well‐understood, routine, and conventional functions… i. Receiving or transmitting data over a network…iv. Storing and retrieving information in memory”) (citing OIP Techs., Inc., v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1093 (Fed. Cir. 2015)). Therefore, recitation of “providing the text chunk as input to an embedding mode” is the well-understood, routine, conventional activity of receiving or transmitting data over a network, as discussed in MPEP § 2106.05(d).
As to claim 15
Step 2A Prong One Analysis: Does the claim recite an abstract idea, law of nature, or natural phenomenon? See MPEP § 2106.04(II)(A)(1).
The limitation “wherein the retrieving further comprises performing a second similarity match between the embedding of the natural language query and the associated embeddings of the second set of text chunks, the performing of the second similarity match resulting in identification of one or more second text chunks in the second set that are relevant to the natural language query; ” is the abstract idea of a mental process that can practically be performed in the human mind, with or without the use of a physical aid such as pen and paper (including an observation, evaluation, judgment, opinion). See MPEP § 2106.04(a)(2)(III).
Step 2A Prong Two Analysis: Does the claim recite additional elements that integrate the judicial exception into a practical application? See MPEP § 2106.04(d).
The limitation.” and placing the one or more second text chunks in the context set” Is an additional element that amounts to adding insignificant extra-solution activity of mere data output and data storing. The claim limitation recites placing the data in context set. See MPEP §§ 2106.05(g)
Step 2B Analysis: Does the claim recite additional elements that amount to significantly more than the judicial exception? See MPEP § 2106.05.
The limitation.” and placing the one or more second text chunks in the context set” Is an additional element that amounts to adding insignificant extra-solution activity of mere data output and data storing. The claim limitation recites placing the data in context set. which the courts have recognized as well‐understood, routine, and conventional when they are claimed in a generic manner. See MPEP §§ 2106.05(g)
As to Claim 16
Step 2A Prong One Analysis: Does the claim recite an abstract idea, law of nature, or natural phenomenon? See MPEP § 2106.04(II)(A)(1).
The limitation “The prompt comprises including the natural language query and the context set in the prompt” is the abstract idea of a mental process that can practically be performed in the human mind, with or without the use of a physical aid such as pen and paper (including an observation, evaluation, judgment, opinion). See MPEP § 2106.04(a)(2)(III).
As to Claim 17
Step 2A Prong One Analysis: Does the claim recite an abstract idea, law of nature, or natural phenomenon? See MPEP § 2106.04(II)(A)(1).
The limitation “and adding the retrieved natural language answer to an LLM prompt that is built for the second natural language query. ” is the abstract idea of a mental process that can practically be performed in the human mind, with or without the use of a physical aid such as pen and paper (including an observation, evaluation, judgment, opinion). See MPEP § 2106.04(a)(2)(III).
Step 2A Prong Two Analysis: Does the claim recite additional elements that integrate the judicial exception into a practical application? See MPEP § 2106.04(d).
The limitation.” storing the natural language answer in a conversational memory database of VSA; and upon receiving a second natural language query from the originator of the natural language query: retrieving the natural language answer from the memory database;” Is an additional element that amounts to adding insignificant extra-solution activity of data storing and data gathering. The claim limitation recites
Storing the answer in memory (i.e data storing) and receiving the data from that database (i.e data gathering). . See MPEP §§ 2106.05(g)
Step 2B Analysis: Does the claim recite additional elements that amount to significantly more than the judicial exception? See MPEP § 2106.05.
The limitation “wherein a plurality of feature attribution values is generated for each feature of a plurality of features of the second machine-learned model” ;” Is an additional element that amounts to adding insignificant extra-solution activity of data storing and data gathering. The claim limitation recites Storing the answer in memory (i.e data storing) and receiving the data from that database (i.e data gathering) which the courts have recognized as well‐understood, routine, and conventional when they are claimed in a generic manner. See MPEP §§ 2106.05(g)
As to Claim 18
Step 1 Analysis: Is the claim to a process, machine, manufacture or composition of matter? See MPEP § 2106.03.
Claim 18 is drawn to a storage medium consisting of Hardware components. i.e (article of manufacture), therefore claim 12 falls under one of four categories of statutory subject matter (machine/products/apparatus, process/method, manufactures and compositions of matter.
Step 2A Prong Two Analysis: Does the claim recite additional elements that integrate the judicial exception into a practical application? See MPEP § 2106.04(d).
The limitation “A non-transitory computer readable storage medium having stored thereon instructions executable by one or more computer systems implementing a virtual solution architect (VSA), the instructions causing the one or more computer systems to” is additional element is an additional element that amounts to adding the words “apply it” (or an equivalent) with the judicial exception or merely uses a computer in its ordinary capacity as a tool to perform an existing process. See MPEP §§ 2106.04(d), 2106.05(f)(2)
Step 2B Analysis: Does the claim recite additional elements that amount to significantly more than the judicial exception? See MPEP § 2106.05.
The limitation “A non-transitory computer readable storage medium having stored thereon instructions executable by one or more computer systems implementing a virtual solution architect (VSA), the instructions causing the one or more computer systems to” is the additional claim elements of one or more processors; a memory coupled to at least one of the processors; a stored instructions stored in the memory and executed by at least one of the processors, are not sufficient to amount to significantly more than the judicial exception since these additional claim elements are recited at a high level of generality (i.e. using a generic processor and generic memory.
And for all the other limitations of Claim 18, it is rejected on the same basis as Claim 1. As the Claim are Analogous.
As to Claim 19
Step 1 Analysis: Is the claim to a process, machine, manufacture or composition of matter? See MPEP § 2106.03.
Claim 19 is a System claim, therefore it falls under one of four categories of statutory subject matter.
Step 2A Prong One Analysis: Does the claim recite an abstract idea, law of nature, or natural phenomenon? See MPEP § 2106.04(II)(A)(1).
The limitation “and identifying one or more business objectives that should be achieved by the technology solution architecture;” is an abstract idea of a mental process that can practically be performed in the human mind, with or without the use of physical aid such as pen and paper (including an observation, evaluation, judgment, opinion). See MPEP § 2106.04(a)(2)(III).
The limitation “build a prompt for a large language model (LLM) using the natural language query and the retrieved portion of the collected knowledge;” is an abstract idea of a mental process that can practically be performed in the human mind, with or without the use of physical aid such as pen and paper (including an observation, evaluation, judgment, opinion). See MPEP § 2106.04(a)(2)(III).
Step 2A Prong Two Analysis: Does the claim recite additional elements that integrate the judicial exception into a practical application? See MPEP § 2106.04(d).
The limitation “A computer system comprising: one or more processors; and a computer readable storage medium having stored thereon program code that, when executed by the one or more processors, cause the one or more processors to” is additional element is an additional element that amounts to adding the words “apply it” (or an equivalent) with the judicial exception or merely uses a computer in its ordinary capacity as a tool to perform an existing process. See MPEP §§ 2106.04(d), 2106.05(f)(2)
The limitation “ collect knowledge relevant to technology solution architecture design” is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering to the judicial exception. The claim recites gathering information from variety of sources See MPEP §§ 2106.04(d), 2106.05(g).
The limitation “ retrieve at least a portion of the collected knowledge based on the natural language query; “is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering to the judicial exception. The claim recites extracting a part of collected information based on the query received. See MPEP §§ 2106.04(d), 2106.05(g).
The limitation “ provide the prompt as input to the LLM, thereby causing the LLM to output a natural language answer to the natural language query answer including the proposed design for the technology solution architecture; “is an additional element that amounts to adding insignificant extra-solution activity of mere data input /output to the judicial exception. The claim limitation recites providing input to model and generating an output based on input(i.e. natural language ).See MPEP §§ 2106.04(d), 2106.05(g).
The limitation “and provide the natural language answer to an originator of the natural language query“ Is an additional element that amounts to adding insignificant extra-solution activity of mere data input /output to the judicial exception. The claim limitation recites providing the output in the form of answer to the user. See MPEP §§ 2106.04(d)
Step 2B Analysis: Does the claim recite additional elements that amount to significantly more than the judicial exception? See MPEP § 2106.05.
The limitation “A non-transitory computer readable storage medium having stored thereon instructions executable by one or more computer systems implementing a virtual solution architect (VSA), the instructions causing the one or more computer systems to” is the additional claim elements of one or more processors; a memory coupled to at least one of the processors; a stored instructions stored in the memory and executed by at least one of the processors, are not sufficient to amount to significantly more than the judicial exception since these additional claim elements are recited at a high level of generality (i.e. using a generic processor and generic memory.
The limitation “collect knowledge relevant to technology solution architecture design” is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering to the judicial exception. The claim recites gathering information from variety of sources which the courts have recognized as well‐understood, routine, and conventional when they are claimed in a generic manner. See MPEP § 2106.05(d)(II).
The limitation “receive a natural language query requesting a proposed design for a technology solution architecture.” is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering to the judicial exception. The claim recites receiving instructions in the form of natural language which the courts have recognized as well‐understood, routine, and conventional when they are claimed in a generic manner. See MPEP §§ 2106.04(d), 2106.05(g).
The limitation “retrieve at least a portion of the collected knowledge based on the natural language query; “is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering to the judicial exception. The claim recites extracting a part of collected information based on the query received which the courts have recognized as well‐understood, routine, and conventional when they are claimed in a generic manner. See MPEP § 2106.05(d)(II).
The limitation “ Provide the prompt as input to the LLM, thereby causing the LLM to output a natural language answer to the natural language query “is an additional element that amounts to adding insignificant extra-solution activity of mere data input /output to the judicial exception. The claim limitation recites providing input to model and generating an output based on input(i.e. natural language ) which the courts have recognized as well‐understood, routine, and conventional when they are claimed in a generic manner. See MPEP § 2106.05(d)(II). See MPEP §§ 2106.04(d), 2106.05(g).
The limitation “the natural language answer including the proposed design for the technology solution architecture; and provide the natural language answer to an originator of the natural language query. “Is an additional element that amounts to adding insignificant extra-solution activity of mere data input /output to the judicial exception. The claim limitation recites providing the output in the form of answer to the user which the courts have recognized as well‐understood, routine, and conventional when they are claimed in a generic manner. See MPEP § 2106.05(d)(II).
Therefore, in examining elements as recited by the limitations individually and as an ordered combination, as a whole the independent claim limitations do not recite what the courts have identified as “significantly more”.
As to Claim 20:
Step 2A Prong Two Analysis: Does the claim recite additional elements that integrate the judicial exception into a practical application? See MPEP § 2106.04(d).
The limitation “a knowledge graph comprising business processes and relationships between each business process and other entities.” is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering to the judicial exception. The claim recites gathering information from variety of sources. See MPEP §§ 2106.04(d), 2106.05(g).
The limitation “a first set of text chunks and associated embeddings, the first set of text chunks originating from application programming interface (API) documentation for a plurality of packaged business capabilities (PBCs).” is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering to the judicial exception. The claim recites gathering information from API. See MPEP §§ 2106.04(d), 2106.05(g).
The limitation “and a second set of text chunks and associated embeddings, the second set of text chunks originating from one or more help websites of one or more vendors of the PBCs. ” is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering to the judicial exception. The claim recites gathering information from internet sites. See MPEP §§ 2106.04(d), 2106.05(g).
Step 2B Analysis: Does the claim recite additional elements that amount to significantly more than the judicial exception? See MPEP § 2106.05.
The limitation “a first set of text chunks and associated embeddings, the first set of text chunks originating from application programming interface (API) documentation for a plurality of packaged business capabilities (PBCs).” is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering to the judicial exception. The claim recites gathering information from API which the courts have recognized as well‐understood, routine, and conventional when they are claimed in a generic manner. See MPEP § 2106.05(d)(II).
The limitation “and a second set of text chunks and associated embeddings, the second set of text chunks originating from one or more help websites of one or more vendors of the PBCs. ” is an additional element that amounts to adding insignificant extra-solution activity of mere data gathering to the judicial exception. The claim recites gathering information from internet sites. which the courts have recognized as well‐understood, routine, and conventional when they are claimed in a generic manner. See MPEP § 2106.05(d)(II).
Therefore, in examining elements as recited by the limitations individually and as an ordered combination, as a whole the independent claim limitations do not recite what the courts have identified as “significantly more”.
Claim Rejections – 35 USC § 103
The following is a quotation of 35 U.S.C. 103, which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
Claim 1-5, 10,18 are rejected under 35 U.S.C. 103 as being unpatentable over Martelaro, et. al., Grant No. US 11720806 B2 (“Martelaro”) in view of Galli et.al Pre-Grant Publication number US 20240289554 A1 (“Galli”) and in view of Qin et. al., Pre-Grant Publication number US 20240346256 A1 (“Qin”)
As To Claim 1
Martelaro teaches “A method performed by one or more computer systems for implementing virtual solution architect (VSA), the method comprising:”(Martelaro, paragraph 14,27,29 “[14] Design recommendation engine 102 [virtual solution architect (VSA)], can be executed on [A method performed by] one or more computing devices [one or more computer systems for], e.g., servers, and can be accessed by one or more users over a network. [26] Design recommendation engine 102 additionally includes an action determination module 120 […] [27] action determination module 120 can receive a user query 124 as input. User query 124 can include a natural language query from a user[…] [29]The action determination module 120 can using natural language processing or other processing techniques to determine [for implementing] a user intent from the one or more terms of the user query 124, where user intent can include one or more component actions or functions.”).
Examiner notes: Under BRI, in light of specification “ paragraph [0011] to address the foregoing and other related issues, the VSA of the present disclosure is a computer-implemented agent that receives natural language queries from users pertaining to the design/creation of a solution architecture and automatically generates natural language answers that are responsive to those queries.” VSA is interpreted as design recommendation engine receiving natural language queries from user and generating response to those queries.( Also see Martelaro [34])
collecting knowledge relevant to technology solution architecture design, the collected knowledge including”(Martelaro “paragraph [26] Design recommendation engine 102 additionally includes an action determination module 120 and design selection module 122, where the actions described herein with reference to the action determination module 120 [ relevant to technology solution architecture design ]and design selection module 122 can be performed by fewer or more modules.[46] In some implementations, the design recommendation engine 102 can utilize the intelligent component design assistant window 204 to collect the additional contextual information 126. [collecting knowledge, the collected knowledge including]The set of context requests 128 provided to the user through the intelligent component design assistant window 204 can be selected using a knowledge graph 132. […] [47] Design recommendation engine 102 can further collect additional contextual information 126 from the user via a requirements list window 206, where the requirements list window 206 presents a set of component design considerations 207 for a user to provide the additional contextual information 126.
a knowledge graph comprising business processes and relationships between each business process and other entities.”(Martelaro, “paragraph [23] The knowledge graph 132 can include [a knowledge graph comprising] nodes each representing a known part(s) with connections to other nodes including functions, use cases, and industries.[ business processes] In one example, a known set of infrared lasers can be linked in the knowledge graph to 1) functions such as LIDAR, free-space data transmittal, and spectroscopy, 2) use cases such as satellites, fiber-optic based data communication, night-vision security, and 3) industrial sectors such as automotive, military, and telecommunications. […][33] As described above, the knowledge graph 132 includes a set of nodes defining components, functions/actions, industries/applications, etc., via a series of connections.[…] [63] The design selection module 122 may identify a set of design recommendations 110 from the database utilizing a knowledge graph 132, where the knowledge graph 132 defines relationships between components, functions, industries/applications, [and relationships between each business process and other entities.] and the like.”).
receiving a natural language query pertaining to a technology solution architecture. (Martelaro, paragraph “[5] In general, one innovative aspect of this technology is directed towards capturing the intelligent context of a design problem posed by the user.[ pertaining to a technology solution architecture] A user can provide a natural language query[receiving a natural language query] to a component recommendation engine, where the query includes terms defining at least one desired function of a component or multiple components, e.g., one or more actions performed by the component.[27]Action determination module 120 can receive a user query 124 as input. User query 124 can include a natural language query from a user, e.g., a designer, where the user query 124 includes one or more terms.”).
Martelaro does not explicitly teach:
a first set of text chunks and associated embeddings, the first set of text chunks originating from application programming interface (API) documentation for a plurality of packaged business capabilities (PBCs).
and a second set of text chunks and associated embeddings, The second set of text chunks originating from one or more help websites of one or more vendors of the packaged business capabilities (PBCs).
retrieving at least a portion of the collected knowledge based on the natural language query. .
building a prompt for a large language model (LLM) using the natural language query and the retrieved portion of the collected knowledge.
providing the prompt as input to the LLM, thereby causing the LLM to output a natural language answer to the natural language query.
and providing the natural language answer to an originator of the natural language query.
Galli teaches “ a first set of text chunks and associated embeddings”(Galli paragraph [0036] As shown in FIG. 1, via embedding 102, the search engine system 100 can pre-calculate vectorized resources 123 [and associated embeddings] based on resources 109. [a first set of text chunks]
Examiner notes: Resource 109 can include one or more textual paragraphs as disclosed by Galli, paragraph [0035]
the first set of text chunks originating from application programming interface (API) documentation for a plurality of packaged business capabilities (PBCs). (Galli “paragraph [0003]For example, a site search can handle questions from a public or private website user, and return answers indexed within or related to a website. In addition, a site search engine can be customized to better suit a business's needs. Furthermore, the site search engine function can be adopted in other areas, such as ticketing support. [for a plurality of packaged business capabilities (PBCs].[0035] According to some embodiments, an answer to a query can be a specific paragraph of a resource. The resources 109 can include all available content, for example, pre-written help articles 110, community posts 112, blogs 114, developer portal 116 [from application programming interface (API) documentation] and academy 118. Furthermore, each source can comprise [originating from] one or more textual paragraphs or sections. [the first set of text chunks]
Under the BRI, in light of specification “[0001] Packaged business capabilities (PBCs) are independent, composable pieces of software that implement business capabilities (or portions thereof) that are part of current business processes. For example, in the domain of electronic commerce, one PBC may implement customer account registration and management, another PBC may implement product catalogs, and yet another PBC may implement shopping cart functionality. Due to their composable nature, PBCs can be combined into a custom technology solution that meets the business needs of a particular organization.” PBC is interpreted as engine better suit a business's needs using technology.
and a second set of text chunks and associated embeddings, The second set of text chunks originating from one or more help websites of one or more vendors of the (PBCs).(Galli, paragraph 0003 0055 “ [0003]For example, a site search can handle questions from a public or private website user, and return answers indexed within or related to a website. In addition, a site search engine can be customized to better suit a business's needs. Furthermore, the site search engine function can be adopted in other areas, such as ticketing support. [one or more vendors of the PBCs]. [0055] A site or domain can include or be associated with resources 202 [the second set of] that could be provided as an answer to a user's question or query. According to some embodiments, an answer can comprise one or more specific paragraphs of a resource. The resources 202 can include all available content, for example, pre-written help articles [of text chunks] community posts, blogs, [originating from one or more help websites] developer portal, and academy. Furthermore, each source can comprise one or more textual paragraphs or sections. [0056] As shown in FIG. 2A, via embedding 210, the search engine system 200 can pre-calculate vectorized resources based on resources 202. [and a second set of text chunks and associated embeddings].
Galli and Martelaro are related to the same field of endeavor (Machine learning Models). In view of the teachings of Galli it would have been obvious for a person of ordinary skill in the art to apply the teachings of Galli to Martelaro before the effective filing date of the claimed invention in order to, collect data form multiple sources as it provides context-rich data pool and embedding those data will result in more accurate and adaptable machine learning models. while responding to the search query. (Galli “[0005] Atter describes improved methods for improved large-scale semantic search and information retrieval via a personalized embedding-based search engine and system. In order to provide a relevant response to a search query, the system can compare the semantic similarities between the content with the consideration of the intent and context of the query.”).
Qin teaches “retrieving at least a portion of the collected knowledge based on the natural language query.”(Qin “ [abs] The first feature vector is compared to a plurality of second feature vectors to determine a subset of the second feature vectors that satisfy a predetermined condition. Augmentation information corresponding to the determined subset of second feature vectors are retrieved. [retrieving at least a portion of the collected knowledge] An augmented prompt, generated based on the query [based on the natural language query]. and the retrieved augmentation information, is provided to a large language model.”).
building a prompt for a large language model (LLM) using the natural language query and the retrieved portion of the collected knowledge. (Qin, paragraph “[0046] Prompt generator 212 may generate a prompt for LLM 214 [building a prompt for a large language model (LLM)] based on one or more of query 216,[ using the natural language query] first feature vector 226, one or more of second feature vectors 228, indications 230, and/or augmentation information 232. [and the retrieved portion of the collected knowledge]”).
providing the prompt as input to the LLM, thereby causing the LLM to output a natural language answer to the natural language query. (Qin “paragraph 0027,0047, “ [0027] The prompt generator provides the augmented prompt with the retrieved augmentation information to the LLM. [providing the prompt as input to the LLM] In an embodiment, the LLM receives the augmented prompt with the contextual information and/or the augmentation information and generates a response [thereby causing the LLM to output] to the original query. [0039]Embodiments, query 216 may include a question in the form of a text string or voice data. [to the natural language query] [0047] For instance, LLM 214 may respond by indicating that it does not know the answer to the query, by generating a response based[ natural language answer to] on the information in its training data, and/or by asking the user to clarify their query.”).
and providing the natural language answer to an originator of the natural language query.(Qin, paragraph “[0102] provide a user interface for querying domain-specific information, wherein the query is received from a user [to an originator of the natural language query.] ( through the user interface; and provide the response to the user through the user interface, [and providing the natural language answer] wherein the response is generated by the large language model based on domain-specific information .”).
Qin and Martelaro are related to the same field of endeavor (Machine learning Models). In view of the teachings of Qin it would have been obvious for a person of ordinary skill in the art to apply the teachings of Qin to Martelaro before the effective filing date of the claimed invention in order to improve accuracy, reduce hallucination and deliver focused, readable response to the user prompt using Machine learning models. (Qin, paragraph “[0018] Embodiments are disclosed herein that improve the scope and accuracy of responses generated by an LLM. […] A retrieval augmented generation (RAG) approach is disclosed herein that adds an information retrieval component to create augmented prompts to feed into the generative language model for generating the final answer/prediction.”).
As to Claim 18
Martelaro teaches “non-transitory computer readable storage medium having stored thereon instructions executable by one or more computer systems implementing a virtual solution architect (VSA), the instructions causing the one or more computer systems to (Martelaro paragraph 14,77 “[14] Design recommendation engine 102 [implementing a virtual solution architect (VSA)], can be executed on one or more computing devices .e.g., servers, and can be accessed by one or more users over a network. [77] In another implementation, the processor 410 is a multi-threaded processor. Processor 410 is capable of processing instructions [the instructions causing the one or more computer systems to] stored in memory 420 or on the storage device 430 [non-transitory computer readable storage medium having stored thereon instructions executable by one or more computer systems] to display graphical information for a user interface on the input/output device 440.
And for all the other limitations of Claim 18, it is rejected under same basis as Claim 1. As the Claim are Analogous.
As To Claim 2
Martelaro in view of Galli and in view of Qin teaches the method of Claim 1.
Martelaro further teaches “wherein the natural language answer comprises a proposed design for the technology solution architecture.” ( Martelaro “paragraph [12] A suggested solution is provided to the querying user [wherein the natural language answer] including at least one recommended component and implementation details. [comprises a proposed design for] The suggested solution can include a set of products, suppliers, etc. Rankings and/or rating-feedback (e.g., star-system) can be provided with the suggested solution, where rankings can identify various filters, e.g., prioritizing different goals, to assist in selecting a solution. Implementation details can include contextual information for form and function of each component included in the solution, e.g., how the component is used, reference designs, technical specifications, identifying limitations/advantages of the design.[ the technology solution architecture]”).
Martelaro, ,Galli and Qin are combinable for the same rationale as set forth above with respect to Claim 1
As to claim 3
Martelaro in view of Galli and in view of Qin teaches the method of Claim 2.
Martelaro further teaches “wherein the natural language query includes one or more business objectives that should be achieved by the technology solution architecture”( Martelaro [27] Action determination module 120 can receive a user query 124 as input. User query 124 can include a natural language query from a user [wherein the natural language query] e.g., a designer, where the user query 124 includes one or more terms. The user query 124 can be descriptive of a design problem including one or more component functions, [includes one or more business objectives that] actions, or operational parameters of an unknown component. […] [29] The action determination module 120 can using natural language processing or other processing techniques to determine[that should be achieved] a user intent from the one or more terms of the user query 124, where user intent can include one or more component actions or functions. [30] The action determination module 120 [by the technology solution architecture]can receive additional contextual information 126 as input in addition to the user query 124. Additional contextual information can provide constraints and/or further specify the desired functions of the desired component(s).
and wherein the proposed design includes a subset of the plurality of PBCs. and a plan for integrating the subset in order to meet one or more business objectives. ( Martelaro paragraph [ 4] a plurality of design recommendations, wherein each design recommendation comprises a respective recommended component and a plurality of recommended component actions, and wherein each of the plurality of recommended component actions is associated with a set of one or more action attributes; selecting,[includes] from the plurality of design recommendations, [and wherein the proposed design] a subset of design recommendations,[ a subset of the plurality of PBCs.] each design recommendation meeting a threshold of similarity between the plurality of the query-derived design functions and the plurality of action attributes for the plurality of query-derived design functions and the plurality of recommended component actions and the action attributes for the recommended components of the design recommendation; and providing, to the user, the subset of design recommendations, [and a plan for integrating the subset in order ]wherein the presentation of each design recommendation comprises the respective recommended component and the plurality of recommended component actions and set of one or more action attributes for the recommended component. [to meet one or more business objectives.]
Martelaro, ,Galli and Qin are combinable for the same rationale as set forth above with respect to Claim 1
As To Claim 4
Martelaro in view of Galli and in view of Qin teaches the method of Claim 1.
Martelaro further teaches “wherein the other entities include one or more PBCs associated with the business process, one or more business domains to which the business process belongs, and one or more industries that the business process is relevant for.” (Martelaro “paragraph [7] The knowledge graph can include a known set of parts with connections to functions, use cases, and industries [wherein the other entities include]. For example, a known set of accelerometers can be linked in the knowledge graph to 1) functions such as motion detection, impact detection, motion tracking, and orientation detection, [one or more PBCs associated with the business process] 2) use cases such as robotics, mobile phones, and flight controllers, and [one or more business domains to which the business process belongs] 3) industrial sectors such as automotive, military, and consumer electronics. [and one or more industries that the business process is relevant for.] An intelligent dialog and/or interactive process, e.g., in the form of a digital assistant, can tailor the conversation with the designers based on the connections in the knowledge graph.
Martelaro, Galli and Qin are combinable for the same rationale as set forth above with respect to Claim 1
As to claim 5
Martelaro in view of Galli and in view of Qin teaches the method of Claim 1.
Galli further teaches “wherein the collected knowledge further includes a third set of text chunks and associated embeddings the third set of text chunks originating from one or more electronic documents.” (Galli, “paragraph [0068]A site or domain [originating from one or more electronic documents.]can be associated with a number of resources 252[ the third set of text chunks,] e.g., resource 1, resource 2 and resource 3. [wherein the collected knowledge further includes] Resource 252 can comprise one or more textual paragraphs or sections.[ a third set of text chunks ] According to some embodiment, either the full content of resource 1 or some selected paragraphs can be provided to query 251 as an answer.[0069] As shown in FIG. 2B, via embedding 254, the search engine system 250 can pre-calculate vectorized resources based on resources 252. [and associated embeddings] For example, sentence transformer 256 can vectorize each resource via partitioning 253 and separately generate a number of source vectors 260.”).
Martelaro, Galli and Qin are combinable for the same rationale as set forth above with respect to Claim 1
As To Claim 10
Martelaro in view of Galli and in view of Qin teaches the method of Claim 1.
Martelaro further teaches “determining a plan for answering the natural language query. “ (Martelaro, “paragraph [003] Implementations of the present disclosure are generally directed towards capturing the context of a design problem posed by the user. [ [determining a plan for answering the natural language query.] A user can provide a natural-language query to a component recommendation engine, where the query includes terms defining at least one desired function of a component or multiple components, e.g., one or more actions performed by the component.”).
and identifying, from the plan, one or more types of solution design knowledge needed to answer the answer the natural language query.“ (Martelaro, “paragraph [4] In some implementations, actions include: receiving, from a user, a user query for a design recommendation, the user query comprising a plurality of terms; determining, from the plurality of terms, [and identifying, from the plan] a user intent including at least one query-derived design function and not including a named component for the design recommendation; receiving, e.g., from the user or otherwise, additional contextual information for the design recommendation, , a subset of design recommendations, each design recommendation meeting a threshold of similarity between the plurality of the query-derived design functions and the plurality of action attributes for the plurality of query-derived design functions and the plurality of recommended component actions and the action attributes for the recommended components of the design recommendation; [one or more types of solution design knowledge needed] and providing, to the user the subset of design recommendations, wherein the presentation of each design recommendation, [answer the answer the natural language query] comprises the respective recommended component and the plurality of recommended component actions and set of one or more action attributes for the recommended component.”).
Martelaro, Galli and Qin are combinable for the same rationale as set forth above with respect to Claim 1
Claim 17 is rejected under 35 U.S.C. 103 as being unpatentable over (“Martelaro”) in view of (“Galli”) and in view of (“Qin”) and in further view of O'Kelly et.al., Grant publication NO. US 11861321 B1 ( “O'Kelly”)
As To Claim 17
Martelaro in view of Galli and in further view of Qin teaches the method of Claim 1.
Martelaro further teaches “storing the natural language answer in a conversational memory database of VSA” (Martelaro, paragraph “[61] Additional contextual information 126 can be user preferences and/or user settings that are automatically or semi-automatically provided to the design recommendation engine 102 from a user profile or tracked user history. In some implementations, the user can opt to have their interactions[ natural language answer in] with the design recommendation engine 102 saved [storing a conversational memory database of VSA.] such that new search queries 124 can be contextualized using past queries, design recommendations, and user feedback.”(.
and upon receiving a second natural language query from the originator of the natural language query: (Martelaro, paragraph 3,6 “[3] A user can provide a natural-language query to a component recommendation engine, [of the natural language query] [6]The method can further include receiving from the user, user feedback responsive to the subset of design recommendations. The user feedback[ and upon receiving a second natural language query from the originator] can include a selection, by the user, of a particular design recommendation of the subset of design recommendations. Collecting user feedback can include collecting usage data for the particular design recommendation including the recommended component.”).
retrieving the natural language answer from the memory database. (Martelaro , paragraph “[61] Additional contextual information 126 can be user preferences and/or user settings that are automatically or semi-automatically provided to the design recommendation engine 102 from a user profile or tracked user history. In some implementations, the user can opt to have their interactions with the design recommendation engine 102 saved[ from the memory database ]such that new search queries 124 can be contextualized using past queries, [retrieving the natural language answer ] design recommendations, and user feedback.”).
Martelaro in view of Galli and in view of Qin does not teach:
and adding the retrieved natural language answer to an LLM prompt that is built for the second natural language query
O'Kelly teaches “and adding the retrieved natural language answer to an LLM prompt that is built for the second natural language query.” (O'Kelly, paragraph 76,82, 84 “[76] In some embodiments, the request may include a query portion. The query portion may include natural language text, one or more instructions in a query language, user input in some other format, or some combination thereof. [82] For instance, all or a portion of the results from an initial request to summarizing a set of text portions may then be used [and adding the retrieved natural language answer ]to create a new set of more compressed input text [that is built for the second natural language query. , which may then be provided [an LLM prompt] to the text generation modeling system for further summarization or other processing.(84) At 410, one or more prompts based on the prompt templates are determined. In some embodiments, a prompt may be determined by supplementing and/or modifying a prompt template based on the input text. For instance, a portion of input text may be added to a prompt template at an appropriate location. As one example, a prompt template may include a set of instructions for causing a large language model to generate a correspondence document. “).
O’Kelly and Martelaro are related to the same field of endeavor (Machine learning Models). In view of the teachings of O’Kelly it would have been obvious for a person of ordinary skill in the art to apply the teachings of O’Kelly to Martelaro before the effective filing date of the claimed invention in order for large language models to tailored accurate response, to user across different domains. (O’Kelly “[23] The structural information and the text portions may then be used to determine a structured document in which the text portions are arranged and organized in association with structural information. Finally, the structured document may be analyzed by a large language model to generate novel text.”).
Claims 6, 11, 13-16 are rejected under 35 U.S.C. 103 as being unpatentable over (“Martelaro”) view of (“Galli”) and in view of (“Qin”) and in view of ( “O'Kelly”) and in further view of Yan et.al, Pre-grant Publication No. US 20200311070 A1 (“Yan”).
As to Claim 6
Martelaro in view of Galli and in view of Qin teaches the method of Claim 1.
Martelaro further teaches “wherein collecting the knowledge comprises: extracting textual data from a knowledge source that holds business process relationship information” (Martelaro, paragraph 20,24 “[20] The recommendation generator module 114 can associate the known parts described in documentation 108 with design information extracted from the documentation 108 including function(s) of the component, the use-case(s), and the relevant industries.[ that holds business process relationship information] [24] In some implementations, the knowledge graph may be populated by using natural language processing tools on the receiving documentation 108 from sources 106. Descriptive text from the various documentation 108 can be extracted [extracting textual data] by the design recommendation generator 104 to identify part features, applications areas, and/or information that references particular design solutions.”).
extracting non-textual data from the knowledge source. (Martelaro, “paragraph [0069] In some implementations, usage data is provided by the recommended component to the design recommendation engine 102, e.g., by an IoT component. [extracting non-textual data from the knowledge source]. For example, a component may be a vibration sensor deployed under water to detect seismic activity. The sensor may provide, via the network, data describing the use and/or utilization of the sensor as feedback to the design recommendation engine 102.”).
O’Kelly further teaches “converting the non-textual data into further textual data” (O’Kelly paragraph 32,72 “[32] The method 300 may be performed in order to convert [converting] a document into [the non-textual data] usable text while at the same time retaining metadata information about the text, [into further textual data ]such as the page, section, and/or document at which the text was located.[72]Optical character recognition is performed on individual pages or on the document as a whole at 308. In some implementations, optical character recognition may be performed locally via a library.
splitting the textual data and the further textual data into a plurality of text chunks;
(O’Kelly paragraph “[96] In some implementations, text [the textual data] may be pre-divided [splitting] into a number of different portions. Examples of divisions of text into portions may include, but are not limited to lists of documents, documents, document sections, document pages, document paragraphs, and document sentences.[ the further textual data into a plurality of text chunks]”).
Martelaro in view of Galli and in view of Qin and in view of O’Kelly does not teach:
and for each text chunk: generating a database insert statement for inserting content of the text chunk into a knowledge graph database storing the knowledge graph
and running the database insert statement on the knowledge graph database
Yan teaches “and for each text chunk: generating a database insert statement for inserting content of the text chunk into a knowledge graph database storing the knowledge graph” (Yan, paragraph 0021, 0023, 0044 “[0021] the memory 14 of the server computing device 10 may be configured to store a knowledge graph 80, [storing the knowledge graph] as discussed in further detail below. The knowledge graph 80 may be a database [a knowledge graph database] of a plurality of ontology entities 82 connected by a plurality of edges to form a graph. Each ontology entity 82 included in the knowledge graph 80 may be a database entry [for inserting content of the text chunk] […] [0023] The processor 12 may be further configured to input the natural language input 20, [0044] Turning now to FIG. 8, which shows additional components of the computing device 10 of FIG. 2, the processor 12 may be further configured to generate a structured query 71, [and for each text chunk: generating a database insert statement for] which may be encoded in a structured query language. Thus, processor 12 may convert the natural language input 20 into a structured query 71 [generating a database insert statement]with which the processor 12 may query the knowledge graph 80.
Examiner notes : Under BRI, inputting a natural language, is interpreted as inserting the text chunk into a knowledge graph. And insert statement under BRI, is interpreted as database structured query for database entry.
and running the database insert statement on the knowledge graph database. (Yan, “paragraph [0021] The memory 14 of the server computing device 10 may be configured to store a knowledge graph 80, as discussed in further detail below. The knowledge graph 80 may be a database [a knowledge graph database] of a plurality of ontology entities 82 connected by a plurality of edges to form a graph. Each ontology entity 82 included in the knowledge graph 80 may be a database entry . When a query of the knowledge graph 80 is performed, [running the database insert statement on]
the results of the query may be one or more output ontology entity tokens 92 included in the knowledge graph 80.
Examiner notes: Under BRI, running a database insert statement is interpreted as running a database structured query which is a database entry.
Yan, and Martelaro are related to the same field of endeavor (Machine learning Models). In view of the teachings of Yan, it would have been obvious for a person of ordinary skill in the art to apply the teachings of Yan, to Martelaro before the effective filing date of the claimed invention in order to, reduce time while improving accuracy during the training of machine learning models to meet user expectation. (Yan, “[0001]Thus, natural language search functionality may allow users without specialized knowledge of query languages to search databases and receive desired search results. Natural language searching may allow the database to be used by a larger number of users and may reduce the amount of time needed to train users of a database.”).
As To Claim 11
Martelaro in view of Galli and in view of Qin teaches the method of Claim 10 .
Martelaro further teaches “upon determining that the one or more type of solution design knowledge include business process relationship information: “ (Martelaro, “paragraph [4] In some implementations, actions include: receiving, from a user, a user query for a design recommendation, the user query comprising a plurality of terms; determining, from the plurality of terms, [upon determining that] a user intent including at least one query-derived design function and not including a named component for the design recommendation; receiving, e.g., from the user or otherwise, additional contextual information for the design recommendation, , a subset of design recommendations, [that the one or more type of solution design knowledge] […]actions and set of one or more action attributes for the recommended component.”). [20] The recommendation generator module 114 can associate the known parts described in documentation 108 with design information extracted from documentation 108 including function(s) of the component, the use-case(s), and the relevant industries.[ include business process relationship information]”).
Yan further teaches “determining, based on the natural language query, a database query statement for querying the business process relationship information in the knowledge graph.”(Yan paragraph “0021, 0022 ,0048 “[0021] The knowledge graph 80 may be a database of a plurality of ontology entities 82 connected by a plurality of edges to form a graph. Each ontology entity 82 included in the knowledge graph 80 may be a database entry. When a query of the knowledge graph 80 is performed, the results of the query may be one or more output ontology entity tokens 92 included in the knowledge graph 80. […] [0022]Using the systems and methods discussed below, processor 12 may determine an output ontology entity token 92 and output a natural language restatement 93 based on the output ontology entity token 92. Thus, the server computing device 10 may respond to the natural language input 20 [determining, based on the natural language query], with an output expressed in natural language. This may allow a user to quickly and easily make a database query. [a database query statement for querying][0048] The first field 97A, “[business.employee],” is filled with the output ontology entity token 92 [the business process relationship information] “Tom Muller.” The second field 97B, “[location.location],” is filled with the input ontology entity token 52 “New York.
Examiner notes: under BRI , based on Natural language query determining a database query statement is interpreted as Yan disclosure of user using a natural language to make a database query. And field 97A filled with output token 92 (described as output of query) with [business employee] interpreted as business process relationship in the knowledge graph.”).
running the database query statement on a knowledge graph database (Yan, paragraph “[0021] The memory 14 of the server computing device 10 may be configured to store a knowledge graph 80, as discussed in further detail below. The knowledge graph 80 may be a database [a knowledge graph database] of a plurality of ontology entities 82 connected by a plurality of edges to form a graph. Each ontology entity 82 included in the knowledge graph 80 may be a database entry . When a query of the knowledge graph 80 is performed, [running the database query statement on] the results of the query may be one or more output ontology entity tokens 92 included in the knowledge graph 80.”).
storing the knowledge graph, thereby causing the knowledge graph database to output the business process relationship information.(Yan, paragraph “[0021] The memory 14 of the server computing device 10 may be configured to store a knowledge graph 80, [storing the knowledge graph] as discussed in further detail below. The knowledge graph 80 may be a database [the knowledge graph database] of a plurality of ontology entities 82 connected by a plurality of edges to form a graph. Each ontology entity 82 included in the knowledge graph 80 may be a database entry. When a query of the knowledge graph 80 is performed, [thereby causing the] the results of the query may be one or more output ontology entity tokens 92 included in the knowledge graph 80. [to output the business process relationship information] [0048] The first field 97A, “[business.employee],” is filled with the output ontology entity token 92 “Tom Muller.” The second field 97B, “[location.location],” is filled with the input ontology entity token 52 “New York.
Examiner notes: fig 9 further details about business process relationship.”).
O’Kelly further teaches “and placing the business process relationship information in a context set for the natural language query.” (O’Kelly, paragraph 243, 289, “[243] In some implementations, the filter prompt may be determined[for the natural language query].by combining the selected one or more portions of the document, [and placing the] some or all of the context information determined at 1704,[ in a context set] the policy identified at 702, a previously generated restatement (i.e., “comprehension”) of the policy generated by the large language model, and a prompt template. The prompt template may include one or more fillable portions in which this information can be inserted. An example of such a prompt template is as follows: Below are portions of two documents. One is our company's policies for contracts. [business process relationship information][…] [289] The natural language instructions may include information such as a format to be used for generating the one or more structural information, an example of structural information to generate, and the like.
Examiner notes: placing the business process relationship is interpreted as combining the document consisting of company’s policies for contract, and context set. under BRI natural language query is interpreted as The filter prompt.”).
Martelaro, Galli, Qin, Yan and O’Kelly are combinable for the same rationale as set forth above with respect to Claim 6
As To Claim 13
Martelaro view of Galli and in view of Qin and in view of Yan and in further view of O’Kelly teaches the method of Claim 11 .
Martelaro further teaches “upon determining that the one or more type of solution design knowledge include other information that is not business process relationship information:(Martelaro, “paragraph 4, 8 [4] In some implementations, actions include: receiving, from a user, a user query for a design recommendation, the user query comprising a plurality of terms; determining, from the plurality of terms, [upon determining that] a user intent including at least one query-derived design function and not including a named component for the design recommendation; receiving, e.g., from the user or otherwise, additional contextual information for the design recommendation, , a subset of design recommendations, [that the one or more type of solution design knowledge] […]actions and set of one or more action attributes for the recommended component.”). [8] A context-informed solution to the design problem can be selected from a structured repository of design recommendations. Each design recommendation can include at least one recommended component and implementation details for the design recommendation. The repository can include documentation provided from multiple sources including, for example, product data sheets, OSHA documentation [include other information that is not business process relationship information], white pages, academic journals, and open source/wiki-based expert contributions.
Qin further teaches “creating an embedding of the natural language query. (Qin paragraph 0049,0050 “[0049] In embodiments, pre-processor 202 generates a query text string 220 based on query 216 and provides query text string 220[of the natural language query.to encoder 204 [0050] In step 304, a first feature vector is generated. For instance , encoder 204 may generate first feature vector 226 that represents the meaning of query text string 220. [creating an embedding of]
Examiner notes: Quin in paragraph [0039] discloses that query 216 includes a question in the form of text string or voice data, and pre-processor 220 generates query text 220 based on the query 216.
performing a first similarity match between the embedding of the natural language query and the associated embeddings of the first set of text chunks, (Qin, Paragraph 0041, 0043 “[0041] In embodiments, encoder 204 may also process augmentation information text string 222 to generate a second feature vector 224 that represents the meaning of augmentation information text string 222. In embodiments, a second feature vector 224 may be generated for each piece of argumentation
information in dataset(s) 112. [0043] In an embodiment, comparator 208 receives first feature vector 226 from encoder 204 [the embedding of the natural language query] and second feature vectors 228 from second feature vectors 206 [and the associated embeddings of the first set of text chunks] and compares first feature vector 226 to second feature vectors 228 to determine the similarity between first feature vector 226 and second feature vectors 228. [performing a first similarity match between ] For example, comparator 208 may calculate a cosine similarity between first feature vector 226 and cash second feature vector 228 to determine second feature vectors that are most similar to first feature vector 226.
the performing of the first similarity match resulting in identification of one or more first text chunks in the first set that are relevant to the natural language query. , (Qin Paragraph “[0044] In embodiments, comparator 208 provides one or more indications 230 to retriever 210. For example, indication(s) 230 may include, but are not limited to, identifiers of second feature vectors that are most similar to first feature vector 226 [the performing of the first similarity match resulting in identification of one or more first text chunks] along with a corresponding cosine similarity score indicating the similarity to first feature vector 226, and/or identifiers of pieces of augmentation information that correspond to the second feature vectors 228 that are most similar to first feature vector 226. [0045] Retriever 210 may be configured to determine and retrieve pieces of argumentation information 232 from dataset(s)112.In embodiments, retriever 210 may receive and analyze indication(s) 230 to identify and retrieve one or more pieces of augmentation information 232 from dataset(s) 112.
Examiner notes: performing the similarity match is interpreted as determining the similarity between first feature vector 226 and second feature vector 228 and resulting in identification of one or more first text chunks in the first set that are relevant to the natural language query is interpreted as identifying one or more pieces of argumentation from dataset.
and placing the one or more first text chunks in the context set. (Qin, paragraph “[0045] Retriever 210 may be configured to determine and retrieve pieces of augmentation information from dataset(s) 112. […] In embodiments, retriever 210 may provide augmentation information 232 [and placing the one or more first text chunks] to prompt generator 212 as part of contextual information 234. [in the context set.]”).
Martelaro, Galli, Qin, Yan and O’Kelly are combinable for the same rationale as set forth above with respect to Claim 11.
As To Claim 14
Martelaro in view of Galli and in view of Qin and in view of Yan and in further view of O’Kelly teaches the method of Claim 13 .
Galli further teaches “wherein the embedding is created by providing the natural language query as input to an embedding model.”(Galli, paragraph 0038 0040,0044 “[0038] According to some embodiments, the embedding 102 can adopt several different sentence transformers [embedding model.], each of which can partition a resource, e.g., a help article, in a different matter. For example, a sentence-transformer model can vectorize all texts in an article, whereas another can only process the first sentence of each paragraph. [0040] Furthermore, query 101 can be a textual query that can be directly provided to [providing the natural language query as input ]embedding 102 for vectorization. Query 101 can also be a question transcribed from voice inputs via a voice-enabled interface. [0044] Upon receiving query 101 or implied query 103, the system can vectorize [wherein the embedding is created by] it via embedding 102 to generate vectorized query 120.
Martelaro, ,Galli ,Qin, Yan and O’Kelly are combinable for the same rationale as set forth above with respect to Claim 11.
As To Claim 16
Martelaro in in view of Galli and in view of Qin and in view of Yan and in further view of O’Kelly teaches the method of Claim 15 .
Qin further teaches “wherein building the prompt comprises including the natural language query and the context set in the prompt.” (Qin, paragraph “[0046] For example, prompt generator 212 may generate an augmented prompt 236 that includes [wherein building the prompt comprises] the original query, [including the natural language query ] contextual information (e.g., the current webpage, the product or service of the current webpage, temporal information, location information, etc.), content information (e.g., the retrieved augmentation information 232), and a request to answer the original query based on the provided contextual information using the included content information.[ and the context set in the prompt.]
Martelaro, ,Galli ,Qin, Yan and O’Kelly are combinable for the same rationale as set forth above with respect to Claim 6
Claims 7 and 12 is rejected under 35 U.S.C. 103 as being unpatentable over (“Martelaro”) in in view of (“Galli”) and in view of (“Qin”) and in view of ( “O'Kelly”) and in view of (“Yan”) and in further view of Cao et.al “A Heterogeneous Graph to Abstract Syntax Tree Framework for Text-to-SQL”(“Cao”)
As To Claim 7
Martelaro in view of Galli and in view of Qin, and in view of Yan and in further view of O’Kelly teaches the method of Claim 6.
Martelaro in view of Galli and in view of Qin, and in view of Yan and in further view of O’Kelly does not teach :
wherein the database insert statement is generated by providing the content of the text chunk and a schema of the knowledge graph database to the LLM.”
Cao teaches “wherein the database insert statement is generated by providing the content of the text chunk and a schema of the knowledge graph database to the LLM.” (Cao [abs]"Text-to-SQL is the task of converting a natural language utterance plus the corresponding database schema into a SQL program." [pg-1, fig1]
PNG
media_image1.png
646
1714
media_image1.png
Greyscale
[abs]"Text-to-SQL is the task of converting a natural language utterance plus the corresponding database schema into a SQL program."
Examiner notes: Under BRI, In light of Specification “[0009] Embodiments of the present disclosure are directed to an automated agent, referred to as a virtual solution architect (VSA), that leverages generative artificial intelligence (AI) (and in particular, generative AI-based large language models (LLMs)) to aid users in designing technology solution architectures via a natural (human) language interface.” Here SQL program is interpreted as database query statement (i.e database insert statement) which is generated by: a question interpreted as text chunk and schema of knowledge graph interpreted as database schema. Text-to-SQL is the task of converting interpreted as process carried out by Large Language model.
Cao and Martelaro are related to the same field of endeavor (Machine learning Models). In view of the teachings of Cao, it would have been obvious for a person of ordinary skill in the art to apply the teachings of Cao, to Martelaro before the effective filing date of the claimed invention in order to, make database more accessible and bridge the gap between human language and structured data while reducing hallucination of Ml models.( Cao [pg-3] To reduce hallucination in grounded language generation [21], both copying [22] and retrieving [23] mechanisms can be adapted to choose chunks in the external memory and insert them at proper places in the output sequence.”)
As To Claim 12
Martelaro in view of Galli and in view of Qin, and in view of Yan and in further view of O’Kelly teaches the method of Claim 11.
Martelaro in view of Galli and in view of Qin, and in view of Yan and in further view of O’Kelly does not teach :
wherein the database query statement is generated by providing the natural language query and a schema of the knowledge graph database to the LLM
Cao teaches “wherein the database query statement is generated by providing the natural language query and a schema of the knowledge graph database to the LLM.” (Cao [abs]" Text-to-SQL is the task of converting a natural language utterance plus the corresponding database schema into a SQL program."
PNG
media_image1.png
646
1714
media_image1.png
Greyscale
Examiner notes: Under BRI, in light of Specification “[0009] Embodiments of the present disclosure are directed to an automated agent, referred to as a virtual solution architect (VSA), that leverages generative artificial intelligence (AI) (and in particular, generative AI-based large language models (LLMs)) to aid users in designing technology solution architectures via a natural (human) language interface.” Here SQL program is interpreted as database query statement which is generated by: a question interpreted as natural language query and schema of knowledge graph interpreted as database schema. Text-to-SQL is the task of converting interpreted as process carried out by LLM.”).
Martelaro, Galli Qin, O’KELLY Yan and Cao are combinable for the same rationale as set forth above with respect to Claim 1
Claims 8 and 9 are rejected under 35 U.S.C. 103 as being unpatentable over (“Martelaro”) in view of (“Galli”) and in view of (“Qin”) and in view of (O’Kelly) and in further view of Lewis et. al “Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks”(“lewis”)
As To Claim 8
Martelaro in view of Galli and in further view of Qin teaches the method of Claim 1.
Galli further teaches “wherein collecting the knowledge comprises: extracting textual data from the API documentation and the one or more help websites.” (Galli 0055 “ [ [0055] A site or domain can include or be associated with resources 202 that could be provided as an answer to a user's question or query. According to some embodiments, an answer can comprise [wherein collecting the knowledge comprises]one or more specific paragraphs of a resource. The resources 202 can include all available content, for example, pre-written help articles community posts, blogs[ and the one or more help websites,] developer portal,[ from the API documentation]and academy. Furthermore, each source can comprise one or more textual paragraphs or sections. [extracting textual data]
Qin further teaches “extracting non-textual data from the API documentation and the one or more help websites.“ (Qin paragraph “[0077] Storage 820 also stores data used and/or generated by operating system 812 and application programs 814 as application data 816. Examples of application data 816 [API documentation] include web pages [the one or more help websites], text, images, tables [non-textual data from the], sound files, video data, and other data, which may also be sent to and/or received from one or more network servers or other devices via one or more wired or wireless networks.”).
O’Kelly further teaches “converting the non-textual data into further textual data.”(O’Kelly paragraph 32, 72 “[32] The method 300 may be performed in order to convert [converting] a document into [the non-textual data] usable text while at the same time retaining metadata information about the text, [into further textual data ]such as the page, section, and/or document at which the text was located.[72]Optical character recognition is performed on individual pages or on the document as a whole at 308. In some implementations, optical character recognition may be performed locally via a library.
Martelaro in view of Galli and in further view of Qin does not teach:
splitting the textual data and the further textual data into a plurality of text chunks; and for each text chunk: creating an embedding of the text chunk; and storing the text chunk and the embedding in a vector database of the VSA.
Lewis teaches “splitting the textual data and the further textual data into a plurality of text chunks; and for each text chunk: creating an embedding of the text chunk; and storing the text chunk and the embedding in a vector database of the VSA.” Lewis [pg-4] “We experiment with RAG in a wide range of knowledge-intensive tasks. For all experiments, we use a single Wikipedia dump for our non-parametric knowledge source. Following Lee et al. [31] and Karpukhin et al. [26], we use the December 2018 dump. Each Wikipedia article is split into disjoint 100-word chunks, to make a total of 21M documents. [splitting the textual data and the further textual data into a plurality of text chunks] We use the document encoder to compute an embedding for each document and build a single MIPS index using FAISS [creating an embedding of the text chunk;] [ pg-3]We refer to the document index as the non-parametric memory. [Pg-2]We build RAG models where the parametric memory is a pre-trained seq2seq transformer, and the non-parametric memory [storing the text chunk and the embedding in a vector database of the VSA] is a dense vector index of Wikipedia, accessed with a pre-trained neural retriever.
Examiner notes: Under BRI, storing the text chunk and embedding in vector database is interpreted as non -parametric memory as it is a vector index of Wikipedia and document index.
Lewis and Martelaro are related to the same field of endeavor (Machine learning Models). In view of the teachings of Lewis, it would have been obvious for a person of ordinary skill in the art to apply the teachings of Lewis to Martelaro before the effective filing date of the claimed invention in order to generate more accurate, specific answers to the user input. (Lewis “[abs] for language generation tasks, we find that RAG models generate more specific, diverse and factual language .”). Additionally, someone who is ordinarily skilled in the art would be motivated to combine the teachings to reduce hallucinations and increase interpretability of ML model as it helps user to be more effective across variety of domains. (Lewis “[pg-10] This work offers several positive societal benefits over previous work: the fact that it is more strongly grounded in real factual knowledge (in this case Wikipedia) makes it “hallucinate” less with generations that are more factual and offers more control and interpretability. RAG could be employed in a wide variety of scenarios with direct benefit to society, for example by endowing it with a medical index and asking it open-domain questions on that topic, or by helping people be more effective at their jobs.”).
As To Claim 9
Martelaro in view of Galli and in view of Qin and in view of O’Kelly and in further view of Lewis teach the method of Claim 8.
Galli further teaches “wherein the embedding is created by providing the text chunk as input to an embedding model.” (Galli, paragraph 0038 0040,0044 “[0038] According to some embodiments, the embedding 102 can adopt several different sentence transformers [embedding model.], each of which can partition a resource, e.g., a help article, in a different matter. For example, a sentence-transformer model can vectorize all texts in an article, whereas another can only process the first sentence of each paragraph. [0040] Furthermore, query 101 can be a textual query that can be directly provided to [providing the text chunk query as input ]embedding 102 for vectorization. Query 101 can also be a question transcribed from voice inputs via a voice-enabled interface. [0044] Upon receiving query 101 or implied query 103, the system can vectorize [wherein the embedding is created by] it via embedding 102 to generate vectorized query 120.
Martelaro , Galli , Qin, O’Kelly and Lewis are combinable for the same rationale as set forth above with respect to Claim 8
As to Claim 15
Martelaro view of Galli and in view of Qin and in view of Yan and in further view of O’Kelly teaches the method of Claim 13 .
Lewis teaches “performing a second similarity match between the embedding of the natural language query and the associated embeddings of the second set of text chunks”
(Lewis [pg-4] Each Wikipedia article is split into disjoint 100-word chunks, [of the second set of text chunks] to make a total of 21M documents. We use the document encoder to compute an embedding for each document [and the associated embeddings,] and build a single MIPS index using FAISS… [pg-3] d(z) is a dense representation of a document produced by a BERTBASE document encoder ,and q(x) a query representation produced by a query encoder [the embedding of the natural language query,] also based on BERTBASE. Calculating top-k(pn(. |x)), the list of k documents z with highest prior probability pn(z|x), is a Maximum Inner Product Search (MIPS) problem, [performing a second similarity match between] which can be approximately solved in sub-linear time.
Examiner notes: Embedding of natural language query is interpreted as q(x) as q(x) is produced by document encoder supplied to the retrieval component pn (z|x). Additionally, performing similarity search is interpreted as MIPS as it is retrieval method of finding similarity match when vectors are normalized.
the performing of the second similarity match resulting in identification of one or more second text chunks in the second set that are relevant to the natural language query(Lewis, pg- 3,4 “[pg-4] Each Wikipedia article is split into disjoint 100-word chunks, to make a total of 21M documents. We use the document encoder to compute an embedding for each document and build a single MIPS index using FAISS with a Hierarchical Navigable Small World approximation for fast retrieval [37]. During training, we retrieve the top k documents [resulting in identification of one or more] for each query.[ that are relevant to the natural language query] [pg-3] d(z) is a dense representation of a document produced by a BERTBASE document encoder, and q(x) a query representation produced by a query also based on BERTBASE. Calculating top-k(pn(. |x)), the list of k documents z[second set of text chunks in the second set that] with highest prior probability pn(z|x), is a Maximum Inner Product Search (MIPS) problem, [performing a second similarity match between] which can be approximately solved in sub-linear time.
Examiner notes: Under BRI, the performing of the second similarity match resulting in identification of one or more second text chunks is interpreted as MIPS search returning top-k of second text chunks of second set z as a result of similarity match and retrieving for each query is interpreted as second set that are relevant to the natural language query.”).
and placing the one or more second text chunks in the context set. (Lewis “[pg-2] [pg-3] We explore RAG models, which use the input sequence x to retrieve text documents z and use them as additional context when generating the target sequence y. As The generator component pθ(yi|x, z, y1:i−1) [the one or more second text chunks]could be modelled using any encoder-decoder. We use BART-large [32], a pre-trained seq2seq transformer [58] with 400M parameters. To combine the input x with the retrieved content z when generating from BART, we simply concatenate them.
Examiner notes: z represents the retrieved second text chunk being used as an additional context by concatenating with input sequence x.( and placing the one or more second text chunks in the context set.] Under BRI , the step as recited above from Lewis is interpreted as placing the one or more second text chunk in context set.”).
Martelaro, ,Galli ,Qin, Yan O’Kelly and Lewis are combinable for the same rationale as set forth above with respect to Claim 8
As To Claim 16
Martelaro in view of Galli and of Qin and in view of Yan and in view of O’Kelly and in further view of Lewis teaches the method of Claim 15 .
Qin further teaches “wherein building the prompt comprises including the natural language query and the context set in the prompt.” (Qin, paragraph “[0046] For example, prompt generator 212 may generate an augmented prompt 236 that includes [wherein building the prompt comprises] the original query, [including the natural language query ] contextual information (e.g., the current webpage, the product or service of the current webpage, temporal information, location information, etc.), content information (e.g., the retrieved augmentation information 232), and a request to answer the original query based on the provided contextual information using the included content information.[ and the context set in the prompt.]
Claim 19 is rejected under 35 U.S.C. 103 as being unpatentable over (“Martelaro”) in view of (“Qin”) .
As to Claim 19
Martelaro teaches “A computer system comprising: one or more processors; and a computer readable storage medium having stored thereon program code that, when executed by the one or more processors, cause the one or more processors to: “ Martelaro, “paragraph [77] in another implementation, the processor 410 is a multi-threaded processor. Processor 410 is capable of processing instructions [when executed by the one or more processors, cause the one or more processors to] stored in memory 420 [A computer system comprising: one or more processors; and a computer readable storage medium having stored thereon program code that] or on the storage device 430 to display graphical information for a user interface on the input/output device 440.
Collect knowledge relevant to technology solution architecture design.(Martelaro “paragraph [26] Design recommendation engine 102 additionally includes an action determination module 120 and design selection module 122, where the actions described herein with reference to the action determination module 120 [ relevant to technology solution architecture design ]and design selection module 122 can be performed by fewer or more modules.[46] In some implementations, the design recommendation engine 102 can utilize the intelligent component design assistant window 204 to collect the additional contextual information 126. [collecting knowledge, the collected knowledge including]The set of context requests 128 provided to the user through the intelligent component design assistant window 204 can be selected using a knowledge graph 132. […] [47] Design recommendation engine 102 can further collect additional contextual information 126 from the user via a requirements list window 206, where the requirements list window 206 presents a set of component design considerations 207 for a user to provide the additional contextual information 126.
receiving a natural language query pertaining to a technology solution architecture. (Martelaro, paragraph “[5] In general, one innovative aspect of this technology is directed towards capturing the intelligent context of a design problem posed by the user.[ pertaining to a technology solution architecture] A user can provide a natural language query[receiving a natural language query] to a component recommendation engine, where the query includes terms defining at least one desired function of a component or multiple components, e.g., one or more actions performed by the component.[27]Action determination module 120 can receive a user query 124 as input. User query 124 can include a natural language query from a user, e.g., a designer, where the user query 124 includes one or more terms.
and identifying one or more business objectives that should be achieved by the technology solution architecture ”( Martelaro, paragraph 27, 29,30 [27] Action determination module 120 can receive a user query 124 as input. User query 124 can include a natural language query from a user e.g., a designer, where the user query 124 includes one or more terms. The user query 124 can be descriptive of a design problem including one or more component functions, [and identifying one or more business objectives that] actions, or operational parameters of an unknown component. […] [29] The action determination module 120 can using natural language processing or other processing techniques to determine[ that should be achieved] a user intent from the one or more terms of the user query 124, where user intent can include one or more component actions or functions. [30] The action determination module 120 [by the technology solution architecture]can receive additional contextual information 126 as input in addition to the user query 124. Additional contextual information can provide constraints and/or further specify the desired functions of the desired component(s).
Examiner notes: Under BRI, one or more business objectives is interpreted as component function and technology solution architecture is interpreted as the action determination module 120 that identifies the business objective from user query that should be achieved accordingly to the user intent.
The natural language answer including the proposed design for the technology solution architecture; ( Martelaro, paragraph [67] the design recommendation engine 102 may receive, from the querying user, user feedback responsive to the provided subset of design recommendations 134, e.g., a user may provide positive or negative reviews of the design choice. In one example, a user may provide a rating for the suggestion design recommendations, for example, via natural language feedback, e.g., “this is not an appropriate part for what I'm looking for.” In another example, a user may provide a rating using a scale, e.g., a number of stars 218. The design recommendation engine 102 may use the received user feedback to refine the knowledge graph 132 and/or rules-based engine for future recommendations responsive to similar user queries 124.
Under BRI, the natural language answer including the proposed design for the technology solution architecture is interpreted as design recommendation engine receiving feedback responsive to the proposed design as design recommendation engine 102 initially provides design recommendation 134 in response of user query( see (Martelaro paragraph[23] [natural language answer including proposed design to solution architecture. Additionally, (Martelaro see paragraph[67] discloses design recommendation engine 102 receives user feedback based on the initially proposed design for technology solution architecture.
Martelaro does not teach:
retrieve at least a portion of the collected knowledge based on the natural language query
build a prompt for a large language model (LLM) using the natural language query and the retrieved portion of the collected knowledge.
provide the prompt as input to the LLM, thereby causing the LLM to output a natural language answer to the natural language query
and provide the natural language answer to an originator of the natural language query
Qin teaches “ retrieve at least a portion of the collected knowledge based on the natural language query”(Qin “ [abs] The first feature vector is compared to a plurality of second feature vectors to determine a subset of the second feature vectors that satisfy a predetermined condition. Augmentation information corresponding to the determined subset of second feature vectors are retrieved. [retrieve at least a portion of the collected knowledge] An augmented prompt, generated based on the query [based on the natural language query]. and the retrieved augmentation information, is provided to a large language model.”).
build a prompt for a large language model (LLM) using the natural language query and the retrieved portion of the collected knowledge. (Qin, paragraph “[0046] Prompt generator 212 may generate a prompt for LLM 214 [build a prompt for a large language model (LLM)] based on one or more of query 216,[ using the natural language query] first feature vector 226, one or more of second feature vectors 228, indications 230, and/or augmentation information 232. [and the retrieved portion of the collected knowledge]”).
provide the prompt as input to the LLM, thereby causing the LLM to output a natural language answer to the natural language query. (Qin “paragraph 0027,0047, “ [0027] The prompt generator provides the augmented prompt with the retrieved augmentation information to the LLM. [provide the prompt as input to the LLM] In an embodiment, the LLM receives the augmented prompt with the contextual information and/or the augmentation information and generates a response [thereby causing the LLM to output] to the original query. [0039]Embodiments, query 216 may include a question in the form of a text string or voice data. [to the natural language query] [0047] For instance, LLM 214 may respond by indicating that it does not know the answer to the query, by generating a response based[ natural language answer to] on the information in its training data, and/or by asking the user to clarify their query.”).
and provide the natural language answer to an originator of the natural language query.(Qin, paragraph “[0102] provide a user interface for querying domain-specific information, wherein the query is received from a user [to an originator of the natural language query.] ( through the user interface; and provide the response to the user through the user interface, [and providing the natural language answer] wherein the response is generated by the large language model based on domain-specific information .”).
Qin and Martelaro are related to the same field of endeavor (Machine learning Models). In view of the teachings of Qin it would have been obvious for a person of ordinary skill in the art to apply the teachings of Qin to Martelaro before the effective filing date of the claimed invention in order to improve accuracy, reduce hallucination and deliver focused, readable response to the user prompt using Machine learning models. (Qin, paragraph “[0018] Embodiments are disclosed herein that improve the scope and accuracy of responses generated by an LLM. […] A retrieval augmented generation (RAG) approach is disclosed herein that adds an information retrieval component to create augmented prompts to feed into the generative language model for generating the final answer/prediction.”).
Claim 20 is rejected under 35 U.S.C. 103 as being unpatentable over (“Martelaro”) in view of (“Qin ”) in further view of (“Galli)”.
As to claim 20
Martelaro in view of Qin teaches the method of System of Claim 19.
Martelaro further teaches “a knowledge graph comprising business processes and relationships between each business process and other entities.”(Martelaro, paragraph “[23] The knowledge graph 132 can include [a knowledge graph comprising] nodes each representing a known part(s) with connections to other nodes including functions, use cases, and industries.[ business processes] In one example, a known set of infrared lasers can be linked in the knowledge graph to 1) functions such as LIDAR, free-space data transmittal, and spectroscopy, 2) use cases such as satellites, fiber-optic based data communication, night-vision security, and 3) industrial sectors such as automotive, military, and telecommunications. […][33] As described above, the knowledge graph 132 includes a set of nodes defining components, functions/actions, industries/applications, etc., via a series of connections.[…] [63] The design selection module 122 may identify a set of design recommendations 110 from the database utilizing a knowledge graph 132, where the knowledge graph 132 defines relationships between components, functions, industries/applications, [and relationships between each business process and other entities.] and the like. “).
Martelaro in view of Qin does not teach.
wherein the collected knowledge includes: a knowledge graph comprising business processes and relationships between each business process and other entities;
a first set of text chunks and associated embeddings, the first set of text chunks originating from application programming interface (API) documentation for a plurality of packaged business capabilities (PBCs);
and a second set of text chunks and associated embeddings, the second set of text chunks originating from one or more help websites of one or more vendors of the PBCs.
Galli teaches “ a first set of text chunks and associated embeddings”(Galli paragraph [0036] As shown in FIG. 1, via embedding 102, the search engine system 100 can pre-calculate vectorized resources 123 [and associated embeddings] based on resources 109. [a first set of text chunks]
Examiner notes: Resource 109 can include one or more textual paragraphs as disclosed by Galli, paragraph [0035]
the first set of text chunks originating from application programming interface (API) documentation for a plurality of packaged business capabilities (PBCs). (Galli “paragraph [0003]For example, a site search can handle questions from a public or private website user, and return answers indexed within or related to a website. In addition, a site search engine can be customized to better suit a business's needs. Furthermore, the site search engine function can be adopted in other areas, such as ticketing support. [one or more vendors of the PBCs].[0035] According to some embodiments, an answer to a query can be a specific paragraph of a resource. The resources 109 can include all available content, for example, pre-written help articles 110, community posts 112, blogs 114, developer portal 116 [from application programming interface (API) documentation] and academy 118. Furthermore, each source can comprise [originating from] one or more textual paragraphs or sections. [the first set of text chunks]
Under the BRI, in light of specification “[0001] Packaged business capabilities (PBCs) are independent, composable pieces of software that implement business capabilities (or portions thereof) that are part of current business processes. For example, in the domain of electronic commerce, one PBC may implement customer account registration and management, another PBC may implement product catalogs, and yet another PBC may implement shopping cart functionality. Due to their composable nature, PBCs can be combined into a custom technology solution that meets the business needs of a particular organization.” PBC is interpreted as engine better suit a business's needs using technology.
and a second set of text chunks and associated embeddings, The second set of text chunks originating from one or more help websites of one or more vendors of the (PBCs).(Galli, paragraph 0003 0055 “ [0003]For example, a site search can handle questions from a public or private website user, and return answers indexed within or related to a website. In addition, a site search engine can be customized to better suit a business's needs. Furthermore, the site search engine function can be adopted in other areas, such as ticketing support. [one or more vendors of the PBCs]. [0055] A site or domain can include or be associated with resources 202 [the second set of] that could be provided as an answer to a user's question or query. According to some embodiments, an answer can comprise one or more specific paragraphs of a resource. The resources 202 can include all available content, for example, pre-written help articles [of text chunks] community posts, blogs, [originating from one or more help websites] developer portal, and academy. Furthermore, each source can comprise one or more textual paragraphs or sections. [0056] As shown in FIG. 2A, via embedding 210, the search engine system 200 can pre-calculate vectorized resources based on resources 202. [and a second set of text chunks and associated embeddings].
Galli and Martelaro are related to the same field of endeavor (Machine learning Models). In view of the teachings of Galli it would have been obvious for a person of ordinary skill in the art to apply the teachings of Galli to Martelaro before the effective filing date of the claimed invention in order to, collect data form multiple sources as it provides context-rich data pool and embedding those data will result in more accurate and adaptable machine learning models. while responding to the search query. (Galli “[0005] Atter describes improved methods for improved large-scale semantic search and information retrieval via a personalized embedding-based search engine and system. In order to provide a relevant response to a search query, the system can compare the semantic similarities between the content with the consideration of the intent and context of the query.”).
Prior Art of Record
The prior art made of record and not relied upon is considered pertinent to applicants’
disclosure.
Bahrami et al. Grant publication No - US 10,853,150 B1 (“Bahrami”) is directed to obtaining API documentation from a knowledge source. Additionally, it also discloses generating database query and executing it on knowledge graph. (Bahrami, “col8, line15-65 .The NLU 320 may thus generate a SPARQL query from the natural language text 310. […]The SPARQL query may then be executed on the knowledge graph 165 and an appropriate API call may be identified in the knowledge graph 165.”).
hull et. al. Grant No. US 11640542 B2 (“Kunde ”) is directed to collecting knowledge associated with architecture recommendation process. (Kunde, paragraph “[11] The recommendation engine 102 also has access to one or more reference databases (not shown in FIG. 1 that provide data which can be used as reference during different stages of data processing associated with the architecture recommendation process.”).
Hull et. al., Grant No US 10846644 B2 (“Hull’) is directed towards the relationship between business process in a knowledge graph and extracting textual and non- textual data from knowledge sources. (Hull, paragraph[65] In one embodiment, process 1000
extracts, for each actionable statement, business entities and metadata related to the actionable statements. […] . Process 1000 may additionally generate the process knowledge graph based on the process fragments and the set of actionable statements and business constraints.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to NIROJ KOIRALA whose telephone number is (571)270-0748. The examiner can normally be reached Monday -Friday 8am-5pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, MICHAEL HUNTLEY can be reached on (303) 297-4307. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/N.K./Examiner, Art Unit 2129
/MICHAEL J HUNTLEY/Supervisory Patent Examiner, Art Unit 2129
Prosecution Insights