# US 12,192,120 B2 · Recommendation based remediation of data insufficiency > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,192,120 B2 - Document ID: US12192120B2 - Status: Granted - Kind: B2 - Application number: 18/389,653 - Filed: 2023-12-19 - Published: 2024-05-23 - Granted: 2025-01-07 - Earliest claimed: 2017-01-27 - Assignees: Triangle IP, Inc. - Applicants: Triangle IP, Inc. - Inventors: Thomas D. Franklin - Web page: https://patentgrep.com/patent/US12192120B2 ## CPC - G06N 20/00: Machine learning - G06F 21/316: {by observing the pattern of computer usage, e.g. typical user behaviour} - G06N 5/04: Inference or reasoning models - G06N 20/20: Ensemble learning - H04L 41/0896: Bandwidth or capacity management, i.e. automatically increasing or decreasing capacities (flow or congestion control using dynamic resource allocation, e.g. in-call renegotiation, H04L47/76) - H04L 41/12: Discovery or management of network topologies - H04L 41/147: for predicting network behaviour - H04L 43/045: for graphical visualisation of monitoring data - H04L 47/83: based on usage prediction - H04L 63/08: {for authentication of entities (cryptographic mechanisms or cryptographic arrangements for entity authentication H04L9/32)} - H04L 41/16: using machine learning or artificial intelligence - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M) ## Abstract The present disclosure describes a patent management system and method for remediating insufficiency of input data for a machine learning system. A plurality of data vectors using data from a plurality of data sources are extracted. A user input with respect to an input data context is received. An input vector based on the user input is generated and a set of matching data vectors are determined from the plurality of data vectors based on the input vector. Data vectors in the set of matching data vectors are determined to be thick data or thin data based on a comparison of a number of matching data vectors with a first pre-determined threshold, and/or a variance with a second pre-determined threshold. Further, the set of matching data vectors are expanded by modifying the input vector when the input data is determined to be insufficient based on a selection of a recommendation. ## Claims ### Claim 1 A method for providing recommendations for remediating insufficiency of input data of a machine learning system, the method comprising: receiving an input data context from a user input; generating an input vector based on the input data context; generating a plurality of data vectors using data extracted from a plurality of data sources, wherein each of the plurality of data vectors comprise a plurality of data elements corresponding to a plurality of dimensions, wherein each of the plurality of dimensions is a property associated with a domain to which the plurality of data vectors belong; generating a set of matching data vectors from among the plurality of data vectors based on the input vector; determining a number of the data vectors included in the set of matching data vectors and a variance from among the plurality of data vectors included in the set of matching data vectors; identifying a first pre-determined threshold and a second pre-determined threshold with respect to the set of matching data vectors; determining whether the data vectors captured in the set of matching data vectors is thin data or thick data with respect to the input data context based on determining at least one of: the number of matching data vectors being less than the first pre-determined threshold, or the variance being less than the second pre-determined threshold; based on the determining of the set of matching data vectors as the thin data, determining the plurality of dimensions for expansion of the set of matching data vectors based on data insufficiency remediation rules; providing a plurality of recommendations to a user for the expansion according to the data insufficiency remediation rules; selecting at least one of the plurality of recommendations according to the data insufficiency remediation rules based on the user input; generating a modified input vector based on the selected at least one of the plurality of recommendations; and generating an expanded set of matching data vectors based on the modified input vector, wherein the input vector is modified by modifying a subset of the plurality of data elements in at least one of the plurality of dimensions. ### Claim 2 (depends on 1) The method for providing recommendations for remediating insufficiency of input data of a machine learning system as recited in claim 1, further comprising determining at least one of the first pre-determined threshold and the second pre-determined threshold by: generating, by a machine learning model, a predicted output based on the set of matching data vectors; determining a degree of correspondence between the predicted output and a corresponding actual output; and determining at least one of the first pre-determined threshold and the second pre-determined threshold, based on the degree of correspondence. ### Claim 3 (depends on 1) The method for providing recommendations for remediating insufficiency of input data of a machine learning system as recited in claim 1, further comprising iteratively adjusting at least one of the first pre-determined threshold and the second pre-determined threshold based on at least one or more data elements included in the modified subset of the plurality of data elements. ### Claim 4 (depends on 1) The method for providing recommendations for remediating insufficiency of input data of a machine learning system as recited in claim 1, wherein the plurality of data sources comprises at least two of: an authenticated data source, an un-authenticated data source, and a private data source. ### Claim 5 (depends on 1) The method for providing recommendations for remediating insufficiency of input data of a machine learning system as recited in claim 1, wherein a recommendation includes diluting the input data context or selecting highest number of data vectors. ### Claim 6 (depends on 1) The method for providing recommendations for remediating insufficiency of input data of a machine learning system as recited in claim 1, wherein the plurality of recommendations are generated by analyzing the plurality of data elements and corresponding dimensions of the set of matching data vectors with respect to data elements of the input vector, and by determining modifications in the data elements of the input vector or modifications in parameters or constraints imposed on the data elements of the input vector. ### Claim 7 (depends on 1) The method for providing recommendations for remediating insufficiency of input data of a machine learning system as recited in claim 1, wherein the data insufficiency remediation rules are to select a recommendation with maximum number of data vectors or select a recommendation with highest variance among the plurality of data vectors. ### Claim 8 A patent management system for providing recommendations for remediating insufficiency of input data of a machine learning system, the patent management system comprising: at least one processor; and at least one memory coupled with the at least one processor, wherein the at least one processor and the at least one memory having instructions are configured to: receive an input data context from a user input; generate an input vector based on the input data context; generate a plurality of data vectors using data extracted from a plurality of data sources, wherein each of the plurality of data vectors comprise a plurality of data elements corresponding to a plurality of dimensions, wherein each of the plurality of dimensions is a property associated with a domain to which the plurality of data vectors belong; generate a set of matching data vectors from among the plurality of data vectors based on the input vector; determine a number of the data vectors included in the set of matching data vectors and a variance from among the plurality of data vectors included in the set of matching data vectors; identify a first pre-determined threshold and a second pre-determined threshold with respect to the set of matching data vectors; determine whether the data vectors captured in the set of matching data vectors is thin data or thick data with respect to the input data context based on determining at least one of: the number of matching data vectors being less than the first pre-determined threshold, or the variance being less than the second pre-determined threshold; based on the determination that the set of matching data vectors is the thin data, determining the plurality of dimensions for expansion of the set of matching data vectors based on data insufficiency remediation rules; provide a plurality of recommendations to a user for the expansion according to the data insufficiency remediation rules; select at least one of the plurality of recommendations according to the data insufficiency remediation rules based on the user input; generate a modified input vector based on the selected at least one of the plurality of recommendations; and generate an expanded set of matching data vectors based on the modified input vector, wherein the input vector is modified by modifying a subset of the plurality of data elements in at least one of the plurality of dimensions. ### Claim 9 (depends on 8) The patent management system for providing recommendations for remediating insufficiency of input data of a machine learning system as recited in claim 8, the patent management system further configured to determine at least one of the first pre-determined threshold and the second pre-determined threshold by: generating, by a machine learning model, a predicted output based on the set of matching data vectors; determining a degree of correspondence between the predicted output and a corresponding actual output; and determining at least one of the first pre-determined threshold and the second pre-determined threshold, based on the degree of correspondence. ### Claim 10 (depends on 8) The patent management system for providing recommendations for remediating insufficiency of input data of a machine learning system as recited in claim 8, patent management system further configured to iteratively adjust at least one of the first pre-determined threshold and the second pre-determined threshold based on at least one or more data elements included in the modified subset of the plurality of data elements. ### Claim 11 (depends on 8) The patent management system for providing recommendations for remediating insufficiency of input data of a machine learning system as recited in claim 8, wherein the plurality of data sources comprises at least two of: an authenticated data source, an un-authenticated data source, and a private data source. ### Claim 12 (depends on 8) The patent management system for providing recommendations for remediating insufficiency of input data of a machine learning system as recited in claim 8, wherein a recommendation includes diluting the input data context or selecting highest number of data vectors. ### Claim 13 (depends on 8) The patent management system for providing recommendations for remediating insufficiency of input data of a machine learning system as recited in claim 8, wherein the plurality of recommendations are generated by analyzing the plurality of data elements and corresponding dimensions of the set of matching data vectors with respect to data elements of the input vector, and by determining modifications in the data elements of the input vector or modifications in parameters or constraints imposed on the data elements of the input vector. ### Claim 14 (depends on 8) The patent management system for providing recommendations for remediating insufficiency of input data of a machine learning system as recited in claim 8, wherein the data insufficiency remediation rules are to select a recommendation with maximum number of data vectors or select a recommendation with highest variance among the plurality of data vectors. ### Claim 15 A patent management system for providing recommendations for remediating insufficiency of input data for a machine learning system, the patent management system comprising: a vector processing server is configured to: generate a plurality of data vectors using data extracted from a plurality of data sources, wherein each of the plurality of data vectors comprise a plurality of data elements corresponding to a plurality of dimensions, wherein each of the plurality of dimensions is a property associated with a domain to which the plurality of data vectors belong; and a thin data processing server configured to: receive an input data context from a user input; generate an input vector based on the input data context; generate a set of matching data vectors from among the plurality of data vectors based on the input vector; determine a number of the data vectors included in the set of matching data vectors and a variance from among the plurality of data vectors included in the set of matching data vectors; identify a first pre-determined threshold and a second pre-determined threshold with respect to the set of matching data vectors; determine whether the data vectors captured in the set of matching data vectors is thin data or thick data with respect to the input data context based on determining at least one of: the number of matching data vectors being less than the first pre-determined threshold, or the variance being less than the second pre-determined threshold; based on the determination that the set of matching data vectors is the thin data, determining the plurality of dimensions for expansion of the set of matching data vectors based on data insufficiency remediation rules; provide a plurality of recommendations to a user for the expansion according to the data insufficiency remediation rules; select at least one of the plurality of recommendations according to the data insufficiency remediation rules based on the user input; generate a modified input vector based on the selected at least one of the plurality of recommendations; and generate an expanded set of matching data vectors based on the modified input vector, wherein the input vector is modified by modifying a subset of the plurality of data elements in at least one of the plurality of dimensions. ### Claim 16 (depends on 15) The patent management system for providing recommendations for remediating insufficiency of input data of a machine learning system as recited in claim 15, the patent management system is further configured to determine at least one of the first pre-determined threshold and the second pre-determined threshold by: generating, by a machine learning model, a predicted output based on the set of matching data vectors; determining a degree of correspondence between the predicted output and a corresponding actual output; and determining at least one of the first pre-determined threshold and the second pre-determined threshold, based on the degree of correspondence. ### Claim 17 (depends on 15) The patent management system for providing recommendations for remediating insufficiency of input data of a machine learning system as recited in claim 15, the patent management system is further configured to iteratively adjust at least one of the first pre-determined threshold and the second pre-determined threshold based on at least one or more data elements included in the modified subset of the plurality of data elements. ### Claim 18 (depends on 15) The patent management system for providing recommendations for remediating insufficiency of input data of a machine learning system as recited in claim 15, wherein the plurality of data sources comprises at least two of: an authenticated data source, an un-authenticated data source, and a private data source. ### Claim 19 (depends on 15) The patent management system for providing recommendations for remediating insufficiency of input data of a machine learning system as recited in claim 15, wherein a recommendation includes diluting the input data context or selecting highest number of data vectors. ### Claim 20 (depends on 15) The patent management system for providing recommendations for remediating insufficiency of input data of a machine learning system as recited in claim 15, wherein the plurality of recommendations are generated by analyzing the plurality of data elements and corresponding dimensions of the set of matching data vectors with respect to data elements of the input vector, and by determining modifications in the data elements of the input vector or modifications in parameters or constraints imposed on the data elements of the input vector.