# US 10,311,058 B1 · Techniques for processing neural queries > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 10,311,058 B1 - Document ID: US10311058B1 - Status: Granted - Kind: B1 - Application number: 16/029,087 - Filed: 2018-07-06 - Granted: 2019-06-04 - Earliest claimed: 2018-07-06 - Assignees: Global Elmeast Inc. - Applicants: Global Elmeast Inc. - Inventors: Manoj Prasanna Kumar - Web page: https://patentgrep.com/patent/US10311058B1 ## CPC - G06F 16/2455: Query execution - G06F 16/211: {Schema design and management} - G06F 16/2458: Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries - G06N 3/042: Knowledge-based neural networks; Logical representations of neural networks - G06N 3/082: modifying the architecture, e.g. adding, deleting or silencing nodes or connections - G06N 3/0895: Weakly supervised learning, e.g. semi-supervised or self-supervised learning - G06N 3/09: Supervised learning - G06N 5/02: Knowledge representation; Symbolic representation - G06N 20/00: Machine learning - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract Approaches, techniques, and mechanisms are disclosed for generating, enhancing, applying and updating knowledge neurons for providing decision making information to a wide variety of client applications. Domain keywords for knowledge domains are generated from domain data of selected domain data sources, along with keyword values for the domain keywords, and are used to generate knowledge artifacts for inclusion in knowledge neurons. These knowledge neurons may be enhanced by domain knowledge data sets found in various data sources and used to generate neural responses to neural queries received from the client applications. Neural feedbacks may be used to update and/or generate knowledge neurons. Any ML algorithm can use, or operate in conjunction with, a neural knowledge artifactory comprising the knowledge neurons to enhance or improve baseline accuracy, for example during a cold start period, for augmented decision making and/or for labeling data points or establishing ground truth to perform supervised learning. ## Claims ### Claim 1 A computer-implemented method comprising: receiving a neural query request for querying a repository of knowledge neurons, wherein the repository of knowledge neurons stores a collection of knowledge neurons obtained based on documents retrieved from a plurality of web-based data sources and knowledge artifacts learned from the documents through machine learning with a machine learning model implemented by a computing device; generating, with a query processor in the computing device, one or more query keywords from the neural query request; searching, based at least in part on a global neural schema and the one or more query keywords generated from the neural query request, for one or more knowledge neurons in the repository of knowledge neurons, wherein the global neural schema comprises parent nodes identifying knowledge domains, child nodes identifying knowledge neurons, and edges relating the parent nodes identifying the knowledge domains to the child nodes identifying the knowledge neurons, wherein the edges specify a set of domain-specific keywords with which the one or more query keywords are to be compared for matching while searching for the one or more knowledge neurons in the repository of knowledge neurons; generating, based at least in part on one or more knowledge artifacts found in the one or more knowledge neurons, a neural query response to the neural query request. ### Claim 2 (depends on 1) The method of claim 1, wherein the neural query request includes a subject profile, and wherein the one or more query keywords comprises one or more subject keywords identified in the subject profile. ### Claim 3 (depends on 1) The method of claim 1, wherein the neural query request includes a predicate profile, and wherein the neural query response specifies a predicted value for a predicate property specified in the predicate profile. ### Claim 4 (depends on 1) The method of claim 1, wherein the global neural schema includes one or more nodes respectively representing one or more knowledge domains, wherein the global neural schema includes one or more child nodes under the one or more nodes representing the one or more knowledge domains, and wherein each child node in the one or more child nodes specifies one or more groups of knowledge neurons in a knowledge domain represented by a parent node of each such child node. ### Claim 5 (depends on 4) The method of claim 4, wherein the global neural schema includes one or more first edges respectively connecting a root node to the one or more nodes, wherein the global neural schema includes one or more second edges respectively connecting the one or more child nodes from one or more parent node of the one or more child nodes. ### Claim 6 (depends on 5) The method of claim 5, wherein each of the one or more first edges specifies a set of knowledge domain keywords for a respective knowledge domain in the one or more knowledge domains, wherein each of the one or more second edges specifies one or more knowledge domain keywords related to knowledge neurons identified by a respective child node in the one or more child nodes. ### Claim 7 (depends on 1) The method of claim 1, wherein the neural query response specifies a predicted value for a property identified in the neural query request, and wherein the predicted value for the property is derived from combining information provided by knowledge artifacts in two or more knowledge neurons. ### Claim 8 A non-transitory computer readable medium that stores computer instructions which, when executed by one or more computing processors, cause the one or more computing processors to perform: receiving a neural query request for querying a repository of knowledge neurons, wherein the repository of knowledge neurons stores a collection of knowledge neurons obtained based on documents retrieved from a plurality of web-based data sources and knowledge artifacts learned from the documents through machine learning with a machine learning model implemented by a computing device; generating, with a query processor in the computing device, one or more query keywords from the neural query request; searching, based at least in part on a global neural schema and the one or more query keywords generated from the neural query request, for one or more knowledge neurons in the repository of knowledge neurons, wherein the global neural schema comprises parent nodes identifying knowledge domains, child nodes identifying knowledge neurons, and edges relating the parent nodes identifying the knowledge domains to the child nodes identifying the knowledge neurons, wherein the edges specify a set of domain-specific keywords with which the one or more query keywords are to be compared for matching while searching for the one or more knowledge neurons in the repository of knowledge neurons; generating, based at least in part on one or more knowledge artifacts found in the one or more knowledge neurons, a neural query response to the neural query request. ### Claim 9 (depends on 8) The non-transitory computer readable medium of claim 8, wherein the neural query request includes a subject profile, and wherein the one or more query keywords comprises one or more subject keywords identified in the subject profile. ### Claim 10 (depends on 8) The non-transitory computer readable medium of claim 8, wherein the neural query request includes a predicate profile, and wherein the neural query response specifies a predicted value for a predicate property specified in the predicate profile. ### Claim 11 (depends on 8) The non-transitory computer readable medium of claim 8, wherein the global neural schema includes one or more nodes respectively representing one or more knowledge domains, wherein the global neural schema includes one or more child nodes under the one or more nodes representing the one or more knowledge domains, and wherein each child node in the one or more child nodes specifies one or more groups of knowledge neurons in a knowledge domain represented by a parent node of each such child node. ### Claim 12 (depends on 11) The non-transitory computer readable medium of claim 11, wherein the global neural schema includes one or more first edges respectively connecting a root node to the one or more nodes, wherein the global neural schema includes one or more second edges respectively connecting the one or more child nodes from one or more parent node of the one or more child nodes. ### Claim 13 (depends on 11) The non-transitory computer readable medium of claim 11, wherein each of the one or more first edges specifies a set of knowledge domain keywords for a respective knowledge domain in the one or more knowledge domains, wherein each of the one or more second edges specifies one or more knowledge domain keywords related to knowledge neurons identified by a respective child node in the one or more child nodes. ### Claim 14 (depends on 8) The non-transitory computer readable medium of claim 8, wherein the neural query response specifies a predicted value for a property identified in the neural query request, and wherein the predicted value for the property is derived from combining information provided by knowledge artifacts in two or more knowledge neurons. ### Claim 15 An apparatus, comprising: one or more computing processors; a non-transitory computer readable medium that stores computer instructions which, when executed by the one or more computing processors, cause the one or more computing processors to perform: receiving a neural query request for querying a repository of knowledge neurons, wherein the repository of knowledge neurons stores a collection of knowledge neurons obtained based on documents retrieved from a plurality of web-based data sources and knowledge artifacts learned from the documents through machine learning with a machine learning model implemented by a computing device; generating, with a query processor in the computing device, one or more query keywords from the neural query request; searching, based at least in part on a global neural schema and the one or more query keywords generated from the neural query request, for one or more knowledge neurons in the repository of knowledge neurons, wherein the global neural schema comprises parent nodes identifying knowledge domains, child nodes identifying knowledge neurons, and edges relating the parent nodes identifying the knowledge domains to the child nodes identifying the knowledge neurons, wherein the edges specify a set of domain-specific keywords with which the one or more query keywords are to be compared for matching while searching for the one or more knowledge neurons in the repository of knowledge neurons; generating, based at least in part on one or more knowledge artifacts found in the one or more knowledge neurons, a neural query response to the neural query request. ### Claim 16 (depends on 15) The apparatus of claim 15, wherein the neural query request includes a subject profile, and wherein the one or more query keywords comprises one or more subject keywords identified in the subject profile. ### Claim 17 (depends on 15) The apparatus of claim 15, wherein the neural query request includes a predicate profile, and wherein the neural query response specifies a predicted value for a predicate property specified in the predicate profile. ### Claim 18 (depends on 15) The apparatus of claim 15, wherein the global neural schema includes one or more nodes respectively representing one or more knowledge domains, wherein the global neural schema includes one or more child nodes under the one or more nodes representing the one or more knowledge domains, and wherein each child node in the one or more child nodes specifies one or more groups of knowledge neurons in a knowledge domain represented by a parent node of each such child node. ### Claim 19 (depends on 18) The apparatus of claim 18, wherein the global neural schema includes one or more first edges respectively connecting a root node to the one or more nodes, wherein the global neural schema includes one or more second edges respectively connecting the one or more child nodes from one or more parent node of the one or more child nodes. ### Claim 20 (depends on 18) The apparatus of claim 18, wherein each of the one or more first edges specifies a set of knowledge domain keywords for a respective knowledge domain in the one or more knowledge domains, wherein each of the one or more second edges specifies one or more knowledge domain keywords related to knowledge neurons identified by a respective child node in the one or more child nodes.