# US 12,118,317 B2 · Techniques to add smart device information to machine learning for increased context > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,118,317 B2 - Document ID: US12118317B2 - Status: Granted - Kind: B2 - Application number: 18/204,767 - Filed: 2023-06-01 - Published: 2023-11-30 - Granted: 2024-10-15 - Earliest claimed: 2019-04-18 - Assignees: Capital One Services, LLC - Applicants: Capital One Services, LLC - Inventors: Alan Salimov; Anish Khazane; Omar Florez Choque - Web page: https://patentgrep.com/patent/US12118317B2 ## CPC - G06F 40/30: Semantic analysis - G06F 16/283: {Multi-dimensional databases or data warehouses, e.g. MOLAP or ROLAP} - G06F 40/205: Parsing - G06F 40/35: Discourse or dialogue representation - G06N 3/006: based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO] - G06N 3/0442: characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU] - G06N 3/088: Non-supervised learning, e.g. competitive learning - G06N 3/09: Supervised learning - G06N 20/00: Machine learning - G06N 20/20: Ensemble learning - G06N 3/043: based on fuzzy logic, fuzzy membership or fuzzy inference, e.g. adaptive neuro-fuzzy inference systems [ANFIS] - G06N 3/044: Recurrent networks, e.g. Hopfield networks - G06N 3/045: Combinations of networks - G06N 3/047: Probabilistic or stochastic networks - G06N 3/049: Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs - G06N 5/01: Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound - G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks - G06N 7/023: {Learning or tuning the parameters of a fuzzy system} - G06N 20/10: using kernel methods, e.g. support vector machines [SVM] - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract Disclosed are an apparatus, a system and a non-transitory computer readable medium that implement processing circuitry that receives non-dialog information from a smart device and determines a data type of data in the received non-dialog information. Based on the determined data type, the processing circuitry transforms the received first data using an input from a machine learning algorithm into transformed data. The transformed data is standardized data that is palatable for machine learning algorithms such as those used implemented as chatbots. The standardized transformed data is useful for training multiple different chatbot systems and enables the typically underutilized non-dialog information to be used to as training input to improve context and conversation flow between a chatbot and a user. ## Claims ### Claim 1 An apparatus to perform contextualized information exchange, comprising: a memory storing instructions; and processing circuitry, coupled to the memory, wherein the processing circuitry is operable to execute the instructions that causes the processing circuitry to perform functions, including functions to: receive information from a user; infer an inferred result, based on the information and health information associated with the user via a first chatbot model, the first chatbot model trained based on data, the data comprising multiple different data types captured from a computing device for contextualization of dialog data, stored in a data structure, with a chain of values, in a standardized format for use in training the first chatbot model, including individual probability values indicating a probability of correspondence between each of a plurality of data elements and probability values indicating a probability of a relationship between the data elements having a correspondence, wherein the chain of values is generated with the data elements that have been transformed, wherein each data element is transformed by a transformation that is selected based on the data type associated with the respective data element, wherein the data comprises non-dialog data including the health information captured by a smart device associated with the user; and generate information to exchange with the user based on the inferred result. ### Claim 2 (depends on 1) The apparatus of claim 1, wherein the instructions further cause the processing circuitry to select the first chatbot model based on the information input from the user, the first chatbot model trained to converse in a regional dialect associated with a location of the user. ### Claim 3 (depends on 1) The apparatus of claim 1, wherein the instructions further cause the processing circuitry to select the first chatbot model based on the information input from the user, wherein the information input relates to financial services, the first chatbot model trained to converse about financial services. ### Claim 4 (depends on 1) The apparatus of claim 1, wherein a category of the non-dialog data comprises one of a fitness device, an exercise device, a personal assistance device, a global positioning system device, a smart phone application that provides non-dialog data, a home security system, a home automation device, an audio system, or a combination thereof. ### Claim 5 (depends on 4) The apparatus of claim 4, wherein a determined data type of the non-dialog data is one of: training center extensible markup language, a spreadsheet protocol, a flexible and interoperable data transfer protocol, a comma-separated value, or global positioning system exchange format. ### Claim 6 (depends on 1) The apparatus of claim 1, the first chatbot model to use a context related vocabulary to generate the inferred result. ### Claim 7 (depends on 1) The apparatus of claim 1, the first chatbot model further trained based on dialog data comprising a context related vocabulary to generate the inferred result. ### Claim 8 A method for contextualized information exchange, comprising: receiving information from a user; inferring an inferred result, based on the information and health information associated with the user via a first chatbot model, the first chatbot model trained based on data, the data comprising multiple different data types captured from a computing device for contextualization of dialog data, stored in a data structure, with a chain of values, in a standardized format for use in training the first chatbot model, including individual probability values indicating a probability of correspondence between each of a plurality of data elements and probability values indicating a probability of a relationship between the data elements having a correspondence, wherein the chain of values is generated with the data elements that have been transformed, wherein each data element is transformed by a transformation that is selected based on the data type associated with the respective data element, wherein the data comprises non-dialog data including the health information captured by a smart device associated with the user; and generating information to exchange with the user based on the inferred result. ### Claim 9 (depends on 8) The method of claim 8, further comprising selecting the first chatbot model based on the information input from the user, the first chatbot model trained to converse in a regional dialect associated with a location of the user. ### Claim 10 (depends on 8) The method of claim 8, further comprising selecting the first chatbot model based on the information input from the user, wherein the information input relates to financial services, the first chatbot model trained to converse about financial services. ### Claim 11 (depends on 8) The method of claim 8, wherein a category of the non-dialog data comprises one of a fitness device, an exercise device, a personal assistance device, a global positioning system device, a smart phone application that provides non-dialog data, a home security system, a home automation device, an audio system, or a combination thereof. ### Claim 12 (depends on 11) The method of claim 11, wherein a determined data type of the non-dialog data is one of: training center extensible markup language, a spreadsheet protocol, a flexible and interoperable data transfer protocol, a comma-separated value, or global positioning system exchange format. ### Claim 13 (depends on 8) The method of claim 8, the first chatbot model to use a context related vocabulary to generate the inferred result. ### Claim 14 (depends on 8) The method of claim 8, the first chatbot model further trained based on dialog data comprising a context related vocabulary to generate the inferred result. ### Claim 15 A non-transitory machine-readable medium comprising instructions, wherein execution of the instructions causes processing circuitry to perform functions to: receive information from a user; infer an inferred result, based on the information and health information associated with the user via a first chatbot model, the first chatbot model trained based on data, the data comprising multiple different data types captured from a computing device for contextualization of dialog data, stored in a data structure, with a chain of values, in a standardized format for use in training the first chatbot model, including individual probability values indicating a probability of correspondence between each of a plurality of data elements and probability values indicating a probability of a relationship between the data elements having a correspondence, wherein the chain of values is generated with the data elements that have been transformed, wherein each data element is transformed by a transformation that is selected based on the data type associated with the respective data element, wherein the data comprises non-dialog data including the health information captured by a smart device associated with the user; and generate information to exchange with the user based on the inferred result. ### Claim 16 (depends on 15) The non-transitory machine-readable medium of claim 15, wherein the instructions further cause the processing circuitry to select the first chatbot model based on the information input from the user, the first chatbot model trained to converse in a regional dialect associated with a location of the user. ### Claim 17 (depends on 15) The non-transitory machine-readable medium of claim 15, wherein the instructions further cause the processing circuitry to select the first chatbot model based on the information input from the user, wherein the information input relates to financial services, the first chatbot model trained to converse about financial services. ### Claim 18 (depends on 15) The non-transitory machine-readable medium of claim 15, wherein a category of the non-dialog data comprises one of a fitness device, an exercise device, a personal assistance device, a global positioning system device, a smart phone application that provides non-dialog data, a home security system, a home automation device, an audio system, or a combination thereof. ### Claim 19 (depends on 18) The non-transitory machine-readable medium of claim 18, wherein a determined data type of the non-dialog data is one of: training center extensible markup language, a spreadsheet protocol, a flexible and interoperable data transfer protocol, a comma-separated value, or global positioning system exchange format. ### Claim 20 (depends on 15) The non-transitory machine-readable medium of claim 15, the first chatbot model further trained based on dialog data comprising a context related vocabulary to generate the inferred result.