DETAILED ACTION
Acknowledgements
This Office Action is in response to Applicant’s response/application filed on 05/17/2024.
The Examiner notes that citations to United States Patent Application Publication paragraphs are formatted as [####], #### representing the paragraph number.
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 .
Status of Claims
Claims 1-20 are currently pending and have been examined.
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claim(s) 1-20 is/are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Bender (US 20220005463).
Regarding claim(s) 1, 10, 11, Bender discloses:
a non-transitory computer readable medium having stored thereon instructions ([0117]);
a processing circuitry ([0117]); and
a memory ([0117]), the memory containing instructions that, when executed by the processing circuitry, configure the system to:
apply a language model to text of a set of natural language interactions in order to output a set of domain-specific language (DSL) data, wherein the set of natural language interactions is between a user and at least one other entity, wherein the set of natural language interactions indicates at least one user-defined concept; (By disclosing, training a machine learning model using natural language interactions comprising applying a language model to the text of a set of natural language interactions/queries in order to output a set of domain-specific language data, wherein user defined concepts includes feedback messages or indicators from the user indicating the relevance of documents ([0104], [0096], [0102], [0097], [0146], [0104], [0124]); “some embodiments may use a set of ontologies to recognize the named entity “benzoyl peroxide” in a first ontology as being associated with the named entity “EPIDUO FORTE” of a second ontology graph, where the second ontology graph may be labeled with a domain class value indicated by the user.” ([0239]));
query a knowledge base based on the set of DSL data in order to obtain at least one DSL query result (By disclosing, queries are searched from a structured knowledge fabric usable to provide data in response to queries ([0124]));
integrate the at least one DSL query result with a structured representation of the natural language interactions in order to create at least one contextualized DSL query result (By disclosing, domain-specific language integrates with natural language interactions to produce queries results ([0104], [0124], [0238])); and
train the language model using the at least one contextualized DSL query result (By disclosing, training a machine learning model based on a set of training queries and a corresponding set of training documents that should be retrieved when the system is provided with the set of training queries ([0104])).
Regarding claim(s) 2, 12, Bender discloses:
creating the structured representation of the set of natural language interactions, wherein the structured representation includes a set of fields and corresponding values, wherein the values in the structured representation include values represented in the natural language interactions (By disclosing, “The data may include a set of existing ontology data 211, a set of natural-language text documents 212, or a set of structured data 214. For example, the set of existing ontology data 210 may include an existing knowledge graph structured in an existing ontology data model, such as the unified medical language system (UMLS) metathesaurus (MeSH). The existing knowledge graph may be stored in various ways, such as in a relational data structure, and may be imported into the ontology data repository 230. As further discussed in this disclosure, different data types may be combined to update an ontology data model, such as one stored in an ontology data model record 231. The ontology data model record may store values for record fields such as object categories, relationships between the categories, directional indicators of the relationships, or the like.” ([0053])).
Regarding claim(s) 3, 13, Bender discloses:
creating the knowledge base based on a dataset associated with a set of files, wherein the created knowledge base indicates a plurality of entities and a plurality of relationships between entities of the plurality of entities indicated among the set of files. (By disclosing, creating a structured knowledge based of a knowledge fabric by obtaining documents, tagged media files, data generated from media files, classify the documents and other data into a set of domain-specific categories and generate or otherwise update a set of ontology graphs based on the provided data, and determine relationships between the set of ontology graphs using one or more machine learning operations of an NLP system to construct or otherwise update a set of ontology graphs that includes one or more ontology graphs, where each graph may be specific to a domain or class within the domain ([0125])).
Regarding claim(s) 4, 14, Bender discloses:
wherein the at least one contextualized DSL query result indicates at least one of the plurality of entities. (By disclosing, “Some embodiments may present the results of the AI services 1234 to a UI that includes UI elements enabling a user to provide feedback. For example, some embodiments may display a generated text to a user in a UI. The user may click on a word of the UI and select one or more domains or domain level that should be assigned to the word, where the selection of the word, domains, or domain levels may be sent in a message to a feedback system 1250. The feedback system 1250 may then update the dataset(s) 1206 to include the user-updated assignment of the word to the domains or domain levels, which may also cause the fine tune training function 1220 to update the parameters of the learning model 1210. For example, some embodiments may update a set of named entity recognition operations based on the updated feedback, where the set of named entity recognition operations may be used in a document comparison operation.” ([0188])).
Regarding claim(s) 5, 15, Bender discloses:
updating the knowledge base on at least a portion of the natural language interactions (By disclosing, updating elements of an ontology graph or other elements of a structured knowledge base in response to user feedback ([0124])),
wherein updating the knowledge base includes adding a representation of a new entity indicated in the portion of the natural language interactions. (By disclosing, “some embodiments may use a set of ontologies to recognize the named entity “benzoyl peroxide” in a first ontology as being associated with the named entity “EPIDUO FORTE” of a second ontology graph, where the second ontology graph may be labeled with a domain class value indicated by the user.” ([0239]); “For example, some embodiments may provide a UI that permits a user to highlight the word “Serological” being displayed in the UI element 2110 and indicate that the word should be added to the second ontology graph via interactions with a set of UI elements. After updating the UI to indicate that the word “Serological” should be added to an ontology, a user may interact with the UI element 2132 by clicking on or tapping on the UI element 2132 to send a message that indicates an update to an ontology graph.” ([0346])).
Regarding claim(s) 6, 16, Bender discloses:
wherein the representation of the new entity added to the knowledge base is expressed in the domain-specific language (By disclosing, “For example, some embodiments may provide a UI that permits a user to highlight the word “Serological” being displayed in the UI element 2110 and indicate that the word should be added to the second ontology graph via interactions with a set of UI elements. After updating the UI to indicate that the word “Serological” should be added to an ontology, a user may interact with the UI element 2132 by clicking on or tapping on the UI element 2132 to send a message that indicates an update to an ontology graph.” ([0346]); ontologies are domain-specific ([0194])).
Regarding claim(s) 7, 17, Bender discloses:
wherein the natural language interactions indicate at least one specification for the at least one user-defined concept. (By disclosing, “For example, if a first n-gram sequence is the first phrase “celery is green” and is in a first sentence stating, “the dog does not like the fact that celery is green and crunchy,” some embodiments may provide a UI that displays the first sentence. The first phrase may be visually indicated and distinct from the surrounding text via highlighting, bolding, text font change, or some other visual indicator. Similarly, some embodiments may display tabular data, where rows, columns, or specific entries of the tabular data may be visually indicated in the UI.” ([0285] of Bender)).
Regarding claim(s) 8, 18, Bender discloses:
wherein the natural language interactions include at least one natural language query (By disclosing, “For example, the query may include a natural language query such as “recent advances in health.”” ([0096])).
Regarding claim(s) 9, 19, Bender discloses:
applying the trained language model to at least one electronic document in order to identify at least one of the at least one user-defined concept in the at least one electronic document. (By disclosing, “As further discussed in this disclosure, some embodiments may perform operations to retrieve documents based on a query sent to the computer system 110, where the documents may be selected based on one or more domain indicators associated with the query. In some embodiments, the domain indicator may be provided with a query or determined from the query.” ([0049])).
Regarding claim(s) 20, Bender discloses:
the knowledge base (KB) storing entities, relations, and concepts (By disclosing, “The database(s) 132 may include or otherwise be capable of accessing data stored in a document database 134, an account database 136, or an ontology database 138. As used in this disclosure, a database may refer to various types of data structures, such as a relational database or a non-relational database.” ([0043]); “Such operations may be especially useful in specialized applications where similar concepts may be disclosed in documents at differing levels of domain expertise, differing levels of security classification, or with differing amounts of relevance to subdomains.” ([0041]); “The dataset(s) 1206 may include publicly available “open” datasets, data specific to an account or organization, additional annotations on a document (e.g., user-entered classifications or named entities), or the like.” ([0186]); “A set of ontology graphs organized in a hierarchy may be used as part of an index for a knowledge fabric, which may include a set of documents or other data, the ontology system(s) and indices used to organize the set of documents or other data, or the functions used use the ontology system(s) or indices used to retrieve information from the set of documents or other data.” ([0060]));
a natural language processing (NLP) component configured to translate the at least one natural language query into the at least one DSL query, wherein the natural language processing component includes the natural language processing model (By disclosing, “the query may include a natural language query such as “recent advances in health.”” ([0096]); and “FIG. 13 is a flowchart of an example process by which a domain-specific summarization may be provided based on a query, in accordance with some embodiments of the present techniques.” ([0195], [0124]));
a structured domain-specific language (DSL) component configured to query the knowledge base using the at least one DSL query (By disclosing, “Some embodiments may perform query expansion using the set of ontology graphs, a trained learning system, or the like. Some embodiments may search through related knowledge systems based on the set of ontology graphs to provide additional graph-based relationships corresponding to domain-specific insights or cross-domain insights in response to updates to the set of ontology graphs.” ([0124])); and
a synthesis component configured to integrate the at least one knowledge base query result with the structured representation of the natural language interactions ([0124], [0238], Fig. 15).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
US 11977854 to Tunstall-Pedoe for disclosing
Methods are provided, such as a method of interacting with a large language model (LLM), including the step of a processing system using a structured, machine-readable representation of data that conforms to a machine-readable language, such as a universal language, to provide new context data for the LLM, in order to improve the output, such as continuation text output, generated by the LLM in response to a prompt; and such as a method of interacting with a LLM, including the step of providing continuation data generated by the LLM to a processing system that uses a structured, machine-readable representation of data that conforms to a machine-readable language, such as a universal language, in which the processing system is configured to analyse the continuation output generated by the LLM in response to a prompt to enable an improved version of that continuation output to be provided to a user. Related computer systems are provided.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to DUAN ZHANG whose telephone number is (571)272-4642. The examiner can normally be reached Mon - Fri 10 AM-5 PM.
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, Neha Patel can be reached at 571-270-1492. 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.
/DUAN ZHANG/Primary Examiner, Art Unit 3699
Prosecution Insights