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 § 103
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 (i.e., changing from AIA to pre-AIA ) 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.
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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1, 2 and 5 are rejected under 35 U.S.C. 103 as being unpatentable over Matias; Yossi et al. (US 9213748 B1) hereinafter MATIAS in view of Osuala; Richard Obinna et al. (US 20240111794 A1) hereinafter OSUALA.
Regarding claim 1, MATIAS teaches:
A semantic search method using example sentences and rearrangements which receives a question from a user and outputs result data corresponding to the question, the method comprising: acquiring the question from the user;
“Methods, systems, and apparatus, including computer programs encoded on computer storage media, for identifying related questions for a search query is described. One of the methods includes receiving a search query from a user device; obtaining a plurality of search results for the search query provided by a search engine, wherein each of the search results identifies a respective search result resource; determining one or more respective topic sets for each search result resource, wherein the topic sets for the search result resource are selected from previously submitted search queries that have resulted in users selecting search results identifying the search result resource; selecting related questions from a question database using the topic sets; and transmitting data identifying the related questions to the user device as part of a response to the search query.” (MATIAS [ABSTRACT]).
“The question database 250 includes search queries that have previously been submitted by users to the search system 214 and that have been determined to be in question form. A query can be determined to be in question form if, e.g., the query includes one of a pre-determined set of question terms. The predetermined set of question terms can include one or more of interrogative words, e.g., interrogative determiners, interrogative pronouns, and interrogative pro-adverbs, other function words that are frequently used to ask a question, or punctuation marks, e.g., question marks. As another example, a query can be determined to be in question form if the query matches one of a predetermined set of question query templates, e.g., “why is [X] used,” where [X] is a placeholder for one or more query terms. ...” (MATIAS [Column 4 line 25 to line 38]).
generating a plurality of answers to the question;
“Methods, systems, and apparatus, including computer programs encoded on computer storage media, for identifying related questions for a search query is described. One of the methods includes receiving a search query from a user device; obtaining a plurality of search results for the search query provided by a search engine, wherein each of the search results identifies a respective search result resource; determining one or more respective topic sets for each search result resource, wherein the topic sets for the search result resource are selected from previously submitted search queries that have resulted in users selecting search results identifying the search result resource; selecting related questions from a question database using the topic sets; and transmitting data identifying the related questions to the user device as part of a response to the search query.” (MATIAS [ABSTRACT]).
defining the plurality of answers as a plurality of example sentences for the question;
“FIG. 1 shows an example search results page 100 for a search query 102 “Lichen planus.” The search results page 100 includes two search results 104 and 106 and related questions 108. The search results 104 and 106 and the related questions 108 are generated by a search system in response to the search query 102. The search results 104 and 106 each identify a respective resource and include respective titles 110 and 120 and respective text snippets 112 and 122 that are extracted from the resources identified by the search results. The search system generates the search results 104 and 106 using conventional search techniques.” (MATIAS [Column 2 line 66 to column 3 line 9]).
arranging the similar data for each example sentence according to the rank;
“In implementations where the system has access to data that associates particular question queries with answers, the system can rank the selected questions based at least in part on whether the question is associated with an answer, and, if the question is associated with an answer, on the quality of the answer that is provided for the question. For example, questions with no answer can be demoted in the ranking or questions with answers can be promoted in the ranking.” (MATIAS [Column 5 line 21 to line 29]).
determining a final rank of the similar data for all of the plurality of example sentences according to a rearrangement criterion;
“Depending on the implementation, when a question is replaced by a best variant, the system can assign to the best variant as the number of times the best variant has been submitted to the search engine the number that is assigned to the question query being replaced by the best variant or the number of times the best variant has actually been submitted. Further, in some implementations, after equivalent questions have been removed or best variants have been identified, the system can re-rank the questions and then again perform step 530 to identify additional equivalent questions.” (MATIAS [Column 7 line 21 to line 30]).
rearranging the similar data according to the final rank;
“Depending on the implementation, when a question is replaced by a best variant, the system can assign to the best variant as the number of times the best variant has been submitted to the search engine the number that is assigned to the question query being replaced by the best variant or the number of times the best variant has actually been submitted. Further, in some implementations, after equivalent questions have been removed or best variants have been identified, the system can re-rank the questions and then again perform step 530 to identify additional equivalent questions.” (MATIAS [Column 7 line 21 to line 30]).
determining the result data from the rearranged similar data;
“… ranking the qualified search queries based on a first number of times each query has been submitted or based on a second number of times users have selected a search result identifying the search result resource after submitting each query; and selecting one or more highest-ranked qualified search queries as the topic sets for the search result resource.” (MATIAS [Column 1 line 55 to line 60]).
and outputting the result data to the user.
“A user 202 can use the user device 204 to submit a query 210 to a search system 214. A search engine 230 within the search system 214 performs a search to identify resources matching the query. When the user 202 submits a query 210, the query 210 may be transmitted through the network 212 to the search system 214. The search system 214 includes an index database 222 and the search engine 230. The search system 214 responds to the query 210 by generating search results 228, which are transmitted through the network to the user device 204 for presentation to the user 102, e.g., as a search results web page to be displayed by a web browser running on the user device 204.” (MATIAS [Column 3 line 47 to line 58]).
MATIAS does not explicitly teach, but OSUALA teaches:
obtaining a similarity by embedding vectors between the plurality of example sentences and a plurality of stored data;
“Texts represented by the selected matching text embeddings may be determined in step 309, e.g., by generation or retrieval. For example, texts represented by the selected matching text embeddings may be retrieved in step 309 by the computer system 101 from the respective sources 103.1-L where they are stored. For each j.sup.th task embedding E.sub.Taskj.sup.1, where j varies between 1 and N, a number r.sub.j.sup.1 of texts ... ” (OSUALA [0073]).
“In one example, the search in the text embedding may be performed using a k-nearest neighbor search, so that that the matching text embeddings for a given searched embedding may comprise a number k of text embeddings obtained by the k-nearest neighbor search. Each of the text embeddings found may be associated with a similarity score. …” (OSUALA [0038]).
deriving similar data by searching the plurality of stored data for each example sentence based on the similarity;
“In one example, the search in the text embedding may be performed using a k-nearest neighbor search, so that that the matching text embeddings for a given searched embedding may comprise a number k of text embeddings obtained by the k-nearest neighbor search. Each of the text embeddings found may be associated with a similarity score. …” (OSUALA [0038]).
assigning a rank to the similar data for each example sentence based on the similarity;
“Alternatively, only selected texts of the texts retrieved for these selected task tokens may be used as a query for a next iteration of the present method. The selected texts may be randomly selected or selected based on a ranking, e.g., the first k ranked texts of each selected task token may be selected. The ranking of texts of each task token may be based on the similarity scores of associated text embedding. For example, the search of a generated embedding Y may result in two text embeddings X1 and X2 (search result embeddings) whose respective texts txt1 and txt2 are retrieved or generated. Texts txt1 and txt2 may be ranked based on the similarity between X1 and Y and the similarity between X2 and Y, respectively. ” (OSUALA [0047]).
It would have been obvious to someone of ordinary skill in the art before the effective filling date of the claimed invention to include in the teachings of MATIAS the capability to generate embeddings of the components, compute the similarity and then using the similarity to search and rank similar data. The benefit and motivation of such modification is discussed by OSUALA in the following portion: “… Embeddings representing in the vector space respective texts of the text source may be generated. These embeddings may be referred to as text embeddings. The text embeddings may be advantageous as they may commonly be produced at once, e.g., using a common encoding step, instead of producing them for every task. This may reduce the disk space requirement and improve the search, which can stay in one embedding space instead of multiple embedding spaces, which may increase efficiency in usage. This may allow comparability of embeddings of different token types, and harnessing multi-task learning/training benefit.” (OSUALA [0027]).
Regarding claim 2, the rejection of claim 1 is incorporated, furthermore MATIAS teaches:
The method according to claim 1, wherein the rearrangement criterion comprises an intersection variable representing the number of times the similar data is commonly derived for the plurality of example sentences,
“The system can measure the quality of an answer based on any of a variety of factors. For example, the quality of the answer may be based at least in part on a quality score generated by the search engine for the resource from which the answer is derived. As another example, the quality of the answer may be based in part on a ranking of a search result identifying the resource from which the answer is derived in a ranking of search results generated by the search engine in response to the question being submitted as a search query. As another example, the quality of the answer may be based in part on the length of the answer, i.e., the number of tokens, terms, or characters in the answer. As another example, if multiple answers are available for a given question, the quality of each answer can be based in part on the number or proportion of terms in the answer that are repeated in other answers for the question. ” (MATIAS [Column 5 line 30 to line 45]).
and a rank variable determining the final rank which represents superiority or inferiority among the similar data for all of the plurality of example sentences based on the rank of the similar data for each example sentence,
“The system selects related questions using the topic sets (step 350). Selecting related questions using topic sets is described in more detail below with reference to FIGS. 5 and 6. Once the related questions have been selected, the system ranks the selected questions based at least in part on the number of times each of the related questions has been submitted to the search engine as a search query. ” (MATIAS [Column 5 line 5 to line 11]).
and the determining of the final rank of the similar data comprises deriving two or more ranks for the similar data for two or more example sentences for which the similar data is commonly derived through the intersection variable,
“The system can measure the quality of an answer based on any of a variety of factors. For example, the quality of the answer may be based at least in part on a quality score generated by the search engine for the resource from which the answer is derived. As another example, the quality of the answer may be based in part on a ranking of a search result identifying the resource from which the answer is derived in a ranking of search results generated by the search engine in response to the question being submitted as a search query. As another example, the quality of the answer may be based in part on the length of the answer, i.e., the number of tokens, terms, or characters in the answer. As another example, if multiple answers are available for a given question, the quality of each answer can be based in part on the number or proportion of terms in the answer that are repeated in other answers for the question. ” (MATIAS [Column 5 line 30 to line 45]).
and determining the final rank through the rank variable.
“In implementations where the system has access to data that associates particular question queries with answers, the system can rank the selected questions based at least in part on whether the question is associated with an answer, and, if the question is associated with an answer, on the quality of the answer that is provided for the question. For example, questions with no answer can be demoted in the ranking or questions with answers can be promoted in the ranking. ” (MATIAS [Column 5 line 22 to line 29]).
“The system can measure the quality of an answer based on any of a variety of factors. For example, the quality of the answer may be based at least in part on a quality score generated by the search engine for the resource from which the answer is derived. As another example, the quality of the answer may be based in part on a ranking of a search result identifying the resource from which the answer is derived in a ranking of search results generated by the search engine in response to the question being submitted as a search query. As another example, the quality of the answer may be based in part on the length of the answer, i.e., the number of tokens, terms, or characters in the answer. As another example, if multiple answers are available for a given question, the quality of each answer can be based in part on the number or proportion of terms in the answer that are repeated in other answers for the question. ” (MATIAS [Column 5 line 30 to line 45]).
Regarding claim 5, the rejection of claim 1 is incorporated, furthermore MATIAS teaches:
The method according to claim 1, wherein the generating of the answers comprises generating the answers according to an example sentence variable which represents the number of the plurality of answers to be defined as the plurality of example sentences.
“Alternatively, instead of or in addition to connecting nodes based on common selected resources, the system can submit each question to the search engine as a search query and obtain search results for each question query. The system can then connect the nodes representing any pair of questions for which at least a first threshold number of search results among a pre-determined number of highest-ranked search results for the two questions identify the same resource. For example, the system may connect the nodes representing any pair of questions where at least, e.g., two search results, in, e.g., the ten highest-ranked search results, for one question in the pair identify the same resource as a search result in the same number of highest-ranked search results for the other question in the pair. ” (MATIAS [Column 7 line 64 to column 8 line 10]).
Claim 3 is rejected under 35 U.S.C. 103 as being unpatentable over MATIAS in view of OSUALA in further view of Mundstock; Walter Cristian Bruck et al. (US 20240193616 A1) hereinafter MUNDSTOCK.
Regarding claim 3, the rejection of claim 2 is incorporated, furthermore MATIAS does not teach, but OSUALA teaches:
applying a plurality of different rearrangement criteria to determine the order of the similar data.
“Alternatively, ranking of the texts of each task token may be based on other parameters than the similarity, namely, the ranking may be based on (a) how successful the search result embedding was for past queries, e.g., measured via a user interaction, (b) popularity of the search result embedding, (c) number of and/or distances to available answer embeddings or other task-specific embeddings, (d) correspondence to a user profile, (e) how general or domain-specific the search result embedding query is, (f) a selection function based on another machine learning model (e.g., neural network) which could, for example, evaluate and rank the relevance or correspondence, and (g) how well the content of and/or user intention of the query could be disambiguated.” (OSUALA [0048]).
It would have been obvious to someone of ordinary skill in the art before the effective filling date of the claimed invention to include in the teachings of MATIAS the capability to apply multiple rearrangement criteria to determine the order of similar data. The benefit and motivation of such modification is discussed by OSUALA in the following portion: “In some embodiments, the method further comprising: for each embedding of the set of embeddings: computing similarity scores between the each embedding and the corresponding search result embeddings; ranking based on the similarity scores the determine data objects corresponding to the each embedding: selecting a subset of the determined data objects based on the ranking; wherein the at least part of the determined objects comprises the selected subsets.” (OSUALA [0126]).
MATIAS in view of OSUALA does not teach, but MUNDSTOCK teaches:
The method according to claim 2, wherein the determining of the final rank of the similar data comprises, when the ranks of the similar data are the same,
“In some embodiments, in the event of two item identifiers previously associated with the user having the same similarity to the item identifier entered by the user in step 235, a tie breaker may be performed, for example, using timestamps of when the two item identifiers were previously associated with the user to prioritize the most recently added item, for example.” (MUNDSTOCK [0049]).
It would have been obvious to someone of ordinary skill in the art before the effective filling date of the claimed invention to include in the teachings of MATIAS in view of OSUALA the capability to determine when the ranks of the similar data are the same, keeping in mind that the ranking of the data is based directly on similarity of the portions. The benefit and motivation of such modification is discussed by MUNDSTOCK in the following portion: “… in the event of two item identifiers previously associated with the user having the same similarity to the item identifier entered by the user in step 235, a tie breaker may be performed…” (MUNDSTOCK [0049]).
Claims 4 and 7 are rejected under 35 U.S.C. 103 as being unpatentable over MATIAS in view of OSUALA in further view of Huh; Domingo et al. (US 20190138615 A1) hereinafter HUH.
Regarding claim 4, the rejection of claim 1 is incorporated, furthermore MATIAS in view of OSUALA does not teach, but HUH teaches:
The method according to claim 1, wherein the deriving of the similar data comprises deriving the similar data according to a range variable which determines the number of similar data to be derived.
“In one particular embodiment, CM recommender 151 may be configured to identify concept markers for suggestion based on the documents returned in response to the user query, e.g., the documents returned in the initial search results. In this case, CM recommender 151 may process the documents in the search results and may identify the concept markers assigned to the documents to generate an initial set of concept markers. In some aspects, CM recommender 151 may not process all documents in the search results, but may process a subset of the results based on a ranking of the documents. For example, CM recommender 151 may process the top n results, with n being a predetermined number (e.g., a number between 1 and 100). The processing of the search result documents may yield the concept markers assigned to the documents. ...” (HUH [0037]).
It would have been obvious to someone of ordinary skill in the art before the effective filling date of the claimed invention to include in the teachings of MATIAS in view of OSUALA the capability to determine a range variable (top n) to determine the number of similar data sources related to the question. The benefit and motivation of such modification is discussed by HUH in the following portion: “Data is the lifeblood of the knowledge economy of today's world. There is, however, a large amount of data available in almost any subject matter area. Identifying relevant data from amongst the available data presents a significant challenge. In existing knowledge search systems, natural language queries are often preferred. In such systems, a user may ask a question or questions, and the system attempts to identify relevant information to address the questions. However, in these systems there is often a disconnect between what the user knows, the question that is asked, and the content data. This disconnect can be the result of the user not knowing particular terms of art or missing important concepts in the query which are essential to finding the most relevant answers.” (HUH [0003])
Regarding claim 7, MATIAS teaches:
A semantic search apparatus using example sentences and rearrangements which receives a question from a user and outputs result data corresponding to the question, the apparatus comprising: an inputter configured to acquire the question from the user;
“A user 202 can use the user device 204 to submit a query 210 to a search system 214. A search engine 230 within the search system 214 performs a search to identify resources matching the query. When the user 202 submits a query 210, the query 210 may be transmitted through the network 212 to the search system 214. The search system 214 includes an index database 222 and the search engine 230. The search system 214 responds to the query 210 by generating search results 228, which are transmitted through the network to the user device 204 for presentation to the user 102, e.g., as a search results web page to be displayed by a web browser running on the user device 204.” (MATIAS [Column 3 line 47 to line 58]).
“Methods, systems, and apparatus, including computer programs encoded on computer storage media, for identifying related questions for a search query is described. One of the methods includes receiving a search query from a user device; obtaining a plurality of search results for the search query provided by a search engine, wherein each of the search results identifies a respective search result resource; determining one or more respective topic sets for each search result resource, wherein the topic sets for the search result resource are selected from previously submitted search queries that have resulted in users selecting search results identifying the search result resource; selecting related questions from a question database using the topic sets; and transmitting data identifying the related questions to the user device as part of a response to the search query.” (MATIAS [ABSTRACT]).
a rearranger configured to determine a final rank of the similar data for all of the plurality of example sentences according to a rearrangement criterion, rearrange the similar data according to the final rank, and determine the result data from the rearranged similar data;
“In implementations where the system has access to data that associates particular question queries with answers, the system can rank the selected questions based at least in part on whether the question is associated with an answer, and, if the question is associated with an answer, on the quality of the answer that is provided for the question. For example, questions with no answer can be demoted in the ranking or questions with answers can be promoted in the ranking.” (MATIAS [Column 5 line 21 to line 29]).
“The system can measure the quality of an answer based on any of a variety of factors. For example, the quality of the answer may be based at least in part on a quality score generated by the search engine for the resource from which the answer is derived. As another example, the quality of the answer may be based in part on a ranking of a search result identifying the resource from which the answer is derived in a ranking of search results generated by the search engine in response to the question being submitted as a search query. As another example, the quality of the answer may be based in part on the length of the answer, i.e., the number of tokens, terms, or characters in the answer. As another example, if multiple answers are available for a given question, the quality of each answer can be based in part on the number or proportion of terms in the answer that are repeated in other answers for the question.” (MATIAS [Column 5 line 30 to line 45]).
and an outputter configured to output the result data to the user.
“A user 202 can use the user device 204 to submit a query 210 to a search system 214. A search engine 230 within the search system 214 performs a search to identify resources matching the query. When the user 202 submits a query 210, the query 210 may be transmitted through the network 212 to the search system 214. The search system 214 includes an index database 222 and the search engine 230. The search system 214 responds to the query 210 by generating search results 228, which are transmitted through the network to the user device 204 for presentation to the user 102, e.g., as a search results web page to be displayed by a web browser running on the user device 204.” (MATIAS [Column 3 line 47 to line 58]).
MATIAS does not teach, but OSUALA teaches:
an example-based searcher configured to obtain a similarity by embedding vectors between the plurality of example sentences and a plurality of stored data, derive similar data by searching for the plurality of stored data for each example sentence based on the similarity, assign a rank to the similar data for each example sentence based on the similarity, and arrange the similar data for each example sentence according to the rank;
“In one example, the search in the text embedding may be performed using a k-nearest neighbor search, so that that the matching text embeddings for a given searched embedding may comprise a number k of text embeddings obtained by the k-nearest neighbor search. Each of the text embeddings found may be associated with a similarity score. The similarity score of a text embedding X indicates the similarity between the text embedding X and the generated embedding Y being searched. That is, the text embedding X has been found as a result of searching the generated embedding Y. The similarity score may, for example, be a distance between text embedding X and generated embedding Y. The distance may, for example, be a cosine distance. The similarity score of the text embedding X may be weighted by a given weight (e.g., between 0 and 1) based on the type of the generated embedding Y. For example, if the generated embedding Y represents the query, then the weight may have the highest value, e.g., 1; if the generated embedding Y represents the supplement of the query that is the answer to the query, then the weight may have a smaller value; and so on.” (OSUALA [0038]).
“Alternatively, only selected texts of the texts retrieved for these selected task tokens may be used as a query for a next iteration of the present method. The selected texts may be randomly selected or selected based on a ranking, e.g., the first k ranked texts of each selected task token may be selected. The ranking of texts of each task token may be based on the similarity scores of associated text embedding. For example, the search of a generated embedding Y may result in two text embeddings X1 and X2 (search result embeddings) whose respective texts txt1 and txt2 are retrieved or generated. Texts txt1 and txt2 may be ranked based on the similarity between X1 and Y and the similarity between X2 and Y, respectively. ” (OSUALA [0047]).
It would have been obvious to someone of ordinary skill in the art before the effective filling date of the claimed invention to include in the teachings of MATIAS the capability to generate embeddings of the components, compute the similarity and then using the similarity to search and rank similar data. The benefit and motivation of such modification is discussed by OSUALA in the following portion: “… Embeddings representing in the vector space respective texts of the text source may be generated. These embeddings may be referred to as text embeddings. The text embeddings may be advantageous as they may commonly be produced at once, e.g., using a common encoding step, instead of producing them for every task. This may reduce the disk space requirement and improve the search, which can stay in one embedding space instead of multiple embedding spaces, which may increase efficiency in usage. This may allow comparability of embeddings of different token types, and harnessing multi-task learning/training benefit.” (OSUALA [0027]).
MATIAS in view of OSUALA does not teach, but HUH teaches:
a generative AI model
“In another embodiment, CM assigner 153 may be configured to assign concept markers to documents in a collection based on an index file associated with the collection. SMEs may identify key concepts in a document and may annotate the document's metadata to indicate the identified key concepts associated with particular content of the document. It is noted that this process for identifying key concepts in a document may be implemented manually, using an automatic analysis process, using an artificial intelligence (AI) process, a combination thereof, etc. …” (HUH [0031]).
“After processing the initial documents in the search results to identify concept markers assigned to the documents, CM recommender 151 may apply a machine learning process, e.g., a classifier, to the identified concept markers in order to classify the concept markers as relevant or non-relevant to the query. In embodiments, models for the classifier may be trained using machine learning algorithms (e.g., gradient boosting trees, Logistic regression, Support Vector Machine (SVM), Naïve Bayes, random forests, neural networks, etc.). ” (HUH [0038]).
It would have been obvious to someone of ordinary skill in the art before the effective filling date of the claimed invention to include in the teachings of MATIAS in view of OSUALA the capability to use an AI model. The benefit and motivation of such modification is discussed by HUH in the following portion: “Thus, it should be appreciated that the techniques and systems disclosed herein provide a technical solution to technical problems existing in the conventional industry practice of search systems. Furthermore, the techniques and systems disclosed herein embody a distinct process and a particular implementation that provides an improvement to existing computer systems by providing the computer systems with new capabilities and functionality for leveraging relevantly ranked concept markers in order to identify, refine, and/or re-rank search results to provide to a user more relevant content in the top ranks, which prior art computer systems do not possess.” (HUH [0018]).
Claim 6 is rejected under 35 U.S.C. 103 as being unpatentable over MATIAS in view of OSUALA in further view of Alupului; Mariana et al. (US 9471668 B1) hereinafter ALUPULUI.
Regarding claim 6, the rejection of claim 1 is incorporated, furthermore MATIAS in view of OSUALA does not teach, but ALUPULUI teaches:
The method according to claim 1, wherein the answers are generated to comprise keywords corresponding to the question, and the generating of the answers comprises generating the plurality of answers to correspond to the question while comprising different keywords.
“A computer program product according to another embodiment provides a question-answering service. The service is facilitated by a first generation subsystem of the computer program product that is configured to receive a first question from a user and computer and to generate at least one first answer to the question. A second generation subsystem is configured to generate a second candidate question based at least in-part on keywords from the first question and the content of the first answer or answers.” (ALUPULUI [Column 1 line 54 to line 62]).
“The method compares the question keyword (e.g., aspirin) and excludes these keywords from the second generation question(s) which removes the possibility of having the same question asked again. The compare process of block 78 is realized between the Alchemy API keywords (entities text/type/concept values) and the initial question keywords and synonyms. ” (ALUPULUI [Column 7 line 1 to line 7]).
It would have been obvious to someone of ordinary skill in the art before the effective filling date of the claimed invention to include in the teachings of MATIAS in view of OSUALA the capability to use keywords in the question-answer generation. The benefit and motivation of such modification is discussed by ALUPULUI in the following portion: “(5) Question-answering (QA) systems, such as the Watson™ system and others, provide answers to user input questions by ingesting a large corpus of documentary data, annotating the data, and generally processing the data before-hand to generate structured information from structured and unstructured electronic documents. With a QA system, a user inputted question is received, the QA system parses the question and analyzes the question to determine what is being asked for, and then performs a search of its ingested data from the corpus to identify candidate answers for the user inputted question, determine confidence scores for the candidate answers based on analysis of evidentiary information, and the like.” (ALUPULUI [Column 2 line 14 to line 26]).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to HECTOR J. CRESPO FEBLES whose telephone number is (571)272-4512. The examiner can normally be reached Mon - Fri 7:30 - 5:00.
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, Daniel Washburn can be reached at (571) 272-5551. 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.
/H.J.C./Examiner, Art Unit 2657
/DANIEL C WASHBURN/Supervisory Patent Examiner, Art Unit 2657
Prosecution Insights