# US 12,182,710 B2 · Machine learning models for automated sustainability data source ingestion and processing > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,182,710 B2 - Document ID: US12182710B2 - Status: Granted - Kind: B2 - Application number: 18/543,415 - Filed: 2023-12-18 - Published: 2024-04-11 - Granted: 2024-12-31 - Earliest claimed: 2021-06-18 - Assignees: Sustainli Inc. - Applicants: Sustainli Inc. - Inventors: Shahla Hebets - Web page: https://patentgrep.com/patent/US12182710B2 ## CPC - G06N 3/08: Learning methods - G06N 20/00: Machine learning - G06F 16/383: using metadata automatically derived from the content - G06N 3/044: Recurrent networks, e.g. Hopfield networks - G06N 3/0442: characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU] - G06N 3/09: Supervised learning - G06Q 10/06393: {Score-carding, benchmarking or key performance indicator [KPI] analysis} - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR ## Abstract A computerized method for automated sustainability data source ingestion and processing includes searching multiple data sources according to specified data ingestion criteria to obtain sustainability data associated with at least one target entity, and supplying the target entity and the obtained sustainability data to a machine learning model to generate a sustainability data source valuation output for each of the multiple data sources. The method includes determining, for each of multiple categories and subcategories, a valuation score based at least in part on one or more of the sustainability data source valuation outputs, and an entity sustainability score based at least in part on one or more of the valuation scores. The method includes determining an overall entity sustainability score associated with the at least one target entity according to the determined entity sustainability scores for the multiple categories and subcategories. ## Claims ### Claim 1 A computer system comprising: memory hardware configured to store a machine learning model, entity sustainability source data vector inputs, and computer-executable instructions, wherein each entity sustainability source data vector input includes sustainability data indicative of at least one sustainability practice of an entity; and processor hardware configured to execute the instructions, wherein the instructions include: training the machine learning model with the entity sustainability source data vector inputs to generate a sustainability data source valuation output, wherein the sustainability data source valuation output includes at least one source valuation score that correlates to a sustainability data source including data indicative of sustainability practices of an entity; obtaining at least one target entity and specified data ingestion criteria; searching multiple data sources according to the specified data ingestion criteria to obtain sustainability data associated with the at least one target entity; supplying the at least one target entity and the obtained sustainability data to the machine learning model to generate a sustainability data source valuation output for each of the multiple data sources that correlates to sustainability data obtained from the data source being indicative of sustainability practices of the at least one target entity; obtaining specified category data and valuation criteria, the specified category data including multiple categories and subcategories; determining, for each of the multiple categories and subcategories, a valuation score based at least in part on one or more of the sustainability data source valuation outputs for obtained sustainability data associated with the category or subcategory; obtaining specified category scoring criteria; determining, for each of the multiple categories and subcategories, an entity sustainability score based at least in part on one or more of the valuation scores associated with the category and subcategories; determining an overall entity sustainability score associated with the at least one target entity according to the determined entity sustainability scores for the multiple categories and subcategories; transmitting the determined overall entity sustainability score associated with the at least one target entity to a computing device to facilitate storage of the overall entity sustainability score in a database data structure or display of the overall entity sustainability score on a user interface; determining multiple overall entity sustainability scores associated with multiple target entities including the at least one target entity; generating a ranking of the multiple overall entity sustainability scores; and displaying a comparison of the ranked overall entity sustainability scores on the user interface. ### Claim 2 (depends on 1) The system of claim 1, wherein the instructions further include: supplying the at least one target entity and the obtained sustainability data to a second model to generate a second sustainability data source valuation output for each of the multiple data sources; for each of the multiple data sources: comparing the sustainability data source valuation output generated by the machine learning model to the second sustainability data source valuation output generated by the second model; and assigning one valuation output to the data source according to a result of the comparison. ### Claim 3 (depends on 2) The system of claim 2, wherein assigning the one valuation output according to the result of the comparison includes: assigning the sustainability data source valuation output generated by the machine learning model to the data source when the result of the comparison is a match; and assigning a greater one of the sustainability data source valuation output generated by the machine learning model to the second sustainability data source valuation output generated by the second model to the data source when the result of the comparison is a non-match. ### Claim 4 (depends on 2) The system of claim 2, wherein the instructions further include: supplying the at least one target entity and the obtained sustainability data to a third model to generate a third sustainability data source valuation output for each of the multiple data sources; for each of the multiple data sources: comparing the sustainability data source valuation output generated by the machine learning model to the second sustainability data source valuation output generated by the second model and the third sustainability data source valuation output generated by the third model; and assigning one valuation output to the data source according to a result of the comparison. ### Claim 5 (depends on 2) The system of claim 2, wherein assigning the one valuation output includes: assigning a high valuation score value when the machine learning model or the second model determines the data source is an audit organization or a certification organization; assigning a medium valuation score value when the machine learning model or the second model determines the data source is the at least one target entity; and assigning a low valuation score value when the machine learning model or the second model determines the data source is a general or unknown data source. ### Claim 6 (depends on 2) The system of claim 2, wherein at least one of the machine learning model and the second model includes a keyword detector with sentiment analysis model, an extractive question answering model, or a generative question answering model. ### Claim 7 (depends on 1) The system of claim 1, wherein the instructions further include displaying the determined entity sustainability scores for the multiple categories and subcategories on the user interface. ### Claim 8 (depends on 1) The system of claim 1, wherein the multiple categories include at least four categories. ### Claim 9 (depends on 8) The system of claim 8, wherein the at least four categories include a climate mitigation category, a fair labor category, an animal welfare category, and a land preservation category. ### Claim 10 A computerized method for automated sustainability data source ingestion and processing, the method comprising: training a machine learning model with entity sustainability source data vector inputs to generate a sustainability data source valuation output, wherein the sustainability data source valuation output includes at least one source valuation score that correlates to a sustainability data source including data indicative of sustainability practices of an entity; obtaining at least one target entity and specified data ingestion criteria; searching multiple data sources according to the specified data ingestion criteria to obtain sustainability data associated with the at least one target entity; supplying the at least one target entity and the obtained sustainability data to the machine learning model to generate a sustainability data source valuation output for each of the multiple data sources that correlates to sustainability data obtained from the data source being indicative of sustainability practices of the at least one target entity; obtaining specified category data and valuation criteria, the specified category data including multiple categories and subcategories; determining, for each of the multiple categories and subcategories, a valuation score based at least in part on one or more of the sustainability data source valuation outputs for obtained sustainability data associated with the category or subcategory; obtaining specified category scoring criteria; determining, for each of the multiple categories and subcategories, an entity sustainability score based at least in part on one or more of the valuation scores associated with the category and subcategories; determining an overall entity sustainability score associated with the at least one target entity according to the determined entity sustainability scores for the multiple categories and subcategories; transmitting the determined overall entity sustainability score associated with the at least one target entity to a computing device to facilitate storage of the overall entity sustainability score in a database data structure or display of the overall entity sustainability score on a user interface; determining multiple overall entity sustainability scores associated with multiple target entities including the at least one target entity; generating a ranking of the multiple overall entity sustainability scores; and displaying a comparison of the ranked overall entity sustainability scores on the user interface. ### Claim 11 (depends on 10) The method of claim 10, further comprising: supplying the at least one target entity and the obtained sustainability data to a second model to generate a second sustainability data source valuation output for each of the multiple data sources; for each of the multiple data sources: comparing the sustainability data source valuation output generated by the machine learning model to the second sustainability data source valuation output generated by the second model; and assigning one valuation output to the data source according to a result of the comparison. ### Claim 12 (depends on 11) The method of claim 11, wherein assigning the one valuation output according to the result of the comparison includes: assigning the sustainability data source valuation output generated by the machine learning model to the data source when the result of the comparison is a match; and assigning a greater one of the sustainability data source valuation output generated by the machine learning model to the second sustainability data source valuation output generated by the second model to the data source when the result of the comparison is a non-match. ### Claim 13 (depends on 11) The method of claim 11, further comprising: supplying the at least one target entity and the obtained sustainability data to a third model to generate a third sustainability data source valuation output for each of the multiple data sources; for each of the multiple data sources: comparing the sustainability data source valuation output generated by the machine learning model to the second sustainability data source valuation output generated by the second model and the third sustainability data source valuation output generated by the third model; and assigning one valuation output to the data source according to a result of the comparison. ### Claim 14 (depends on 11) The method of claim 11, wherein assigning the one valuation output includes: assigning a high valuation score value when the machine learning model or the second model determines the data source is an audit organization or a certification organization; assigning a medium valuation score value when the machine learning model or the second model determines the data source is the at least one target entity; and assigning a low valuation score value when the machine learning model or the second model determines the data source is a general or unknown data source. ### Claim 15 (depends on 11) The method of claim 11, wherein at least one of the machine learning model and the second model includes a keyword detector with sentiment analysis model, an extractive question answering model, or a generative question answering model. ### Claim 16 (depends on 10) The method of claim 10, further comprising displaying the determined entity sustainability scores for the multiple categories and subcategories on the user interface. ### Claim 17 (depends on 10) The method of claim 10, wherein the multiple categories include at least four categories. ### Claim 18 (depends on 17) The method of claim 17, wherein the at least four categories include a climate mitigation category, a fair labor category, an animal welfare category, and a land preservation category.