# US 10,872,692 B2 · Data driven analysis, modeling, and semi-supervised machine learning for qualitative and quantitative determinations > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 10,872,692 B2 - Document ID: US10872692B2 - Status: Granted - Kind: B2 - Application number: 16/119,018 - Filed: 2018-08-31 - Published: 2018-12-27 - Granted: 2020-12-22 - Earliest claimed: 2016-06-01 - Assignees: Grand Rounds, Inc. - Applicants: Grand Rounds, Inc. - Inventors: Seiji James Yamamoto; Ranjit Chacko - Web page: https://patentgrep.com/patent/US10872692B2 ## CPC - G16H 40/20: for the management or administration of healthcare resources or facilities, e.g. managing hospital staff or surgery rooms - G06F 16/285: {Clustering or classification} - G06F 16/9024: {Graphs; Linked lists (G06F16/9027 takes precedence)} - G16H 50/20: for computer-aided diagnosis, e.g. based on medical expert systems - G16Z 99/00: Subject matter not provided for in other main groups of this subclass - G06F 16/355: Creation or modification of classes or clusters - G06Q 10/10: Office automation; Time management - G06Q 50/22: Social work or social welfare, e.g. community support activities or counselling services - G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G16Z: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS, NOT OTHERWISE PROVIDED FOR - 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 Systems and methods are provided for data driven analysis, modeling, and semi-supervised machine learning for qualitative and quantitative determinations. The systems and methods include obtaining data associated with individuals, and determining features associated with the individuals based on the data and similarities among the individuals based on the features. The systems and methods can label some individuals as exemplary, generate a graph wherein nodes of the graph represent individuals, edges of the graph represent similarity among the individuals, and nodes associated labeled individuals are weighted. The disclosed system and methods can apply a weight to unweighted nodes of the graph based on propagating the labels through the graph where the propagation is based on influence exerted by the weighted nodes on the unweighted nodes. The disclosed systems and methods can provide output associated with the individuals represented on the graph and the associated weights. ## Claims ### Claim 1 A non-transitory computer readable storage medium storing instructions that are executable by a first computing device that includes one or more processors to cause the first computing device to perform a method comprising: obtaining, from one or more data sources, one or more data sets associated with a plurality of individuals; determining features associated with the plurality of individuals; determining similarities among the plurality of individuals based on the determined features; generating data representing a graph having a plurality of nodes, wherein the nodes of the graph are associated with the plurality of individuals, edges between the nodes represent the determined similarities, wherein the nodes include weighted nodes that are associated with a first subset of the plurality of individuals and unweighted nodes; applying a weight to the unweighted nodes of the graph, wherein the weights are based on an influence exerted by the weighted nodes on the unweighted nodes; providing output associated with the graph and the applied weights. ### Claim 2 (depends on 1) The non-transitory computer readable storage medium of claim 1, wherein the graph is fully connected. ### Claim 3 (depends on 1) The non-transitory computer readable storage medium of claim 1, wherein generating the graph utilizes a random graph model to select a subset of individuals to connect using edges on the graph. ### Claim 4 (depends on 1) The non-transitory computer readable storage medium of claim 1, wherein the edges of the graph are weighted and the weight of an edge is based on the determined similarity between individuals associated with the nodes connected by the edge. ### Claim 5 (depends on 1) The non-transitory computer readable storage medium of claim 1, wherein the first subset of plurality of individuals are labeled as exemplary comprises: one or more individuals of the first subset of individuals that are labeled as positive and one or more individuals of the first subset of individuals that are labeled as negative. ### Claim 6 (depends on 5) The non-transitory computer readable storage medium of claim 5, further comprising: labeling the first subset of the plurality of individuals comprising: labeling a first set of individuals of the plurality of individuals as exemplary, wherein the first set of individuals includes at least one individual that is associated with a subset of features that is correlated to qualities associated with exemplary individuals; identifying a second set of individuals of the plurality of individuals that are similar to at least one individual of the first set, wherein the similarity is based on one or more features associated with the at least one individual and the second set of individuals; and labeling the second set of individuals as exemplary. ### Claim 7 (depends on 1) The non-transitory computer readable storage medium of claim 1, wherein the influence of the weighted nodes on the unweighted nodes is determined using Poisson's equation, Laplace's equation, a Laplacian exponential diffusion kernel, a regularized Laplacian kernel, or a Von Neumann diffusion kernel. ### Claim 8 (depends on 1) The non-transitory computer readable storage medium of claim 1, wherein providing the output further comprises providing the graph and the associated weights for processing by a client device. ### Claim 9 (depends on 1) The non-transitory computer readable storage medium of claim 1, wherein providing output further comprises providing the output for display on a graphical user interface. ### Claim 10 An electronic device comprising: a data storage configured to store one or more data sets associated with a plurality of individuals; a feature discovery engine configured to: determine one or more features of the one or more data sets; a semi-supervised engine configured to: determine similarities among the plurality of individuals, based on the determined features; and generate data representing a graph having a plurality of nodes, wherein the nodes of the graph are associated with the plurality of individuals, edges between the nodes represent the determined similarities, wherein the nodes include weighted nodes that are associated with a first subset of the plurality of individuals and unweighted nodes; a label propagation engine configured to: applying a weight to the unweighted nodes of the graph, wherein the weights are based on an influence exerted by the weighted nodes on the unweighted nodes; and an evaluation engine configured to provide output associated with the graph and the applied weights. ### Claim 11 (depends on 10) The electronic device of claim 10, wherein the semi-supervised engine utilizes a random graph model to select a subset of similar individuals to connect using edges on the graph. ### Claim 12 (depends on 10) The electronic device of claim 10, wherein the edges of the graph are weighted and the weight of an edge is based on the determined similarity between the individuals represented by nodes connected by the edge. ### Claim 13 (depends on 10) The electronic device of claim 10, further comprising: a label engine configured to: identify the first subset of the plurality of individuals as exemplary; and label the first subset as exemplary, wherein one or more individuals of the first subset are labeled as positive and one or more individuals of the first subset are labeled as negative. ### Claim 14 (depends on 13) The electronic device of claim 13, wherein the label engine is configured to: label a first set of individuals of the plurality of individuals as exemplary wherein the first set of individuals includes at least one individual that is associated with a subset of features wherein the subset of features correlate to qualities associated with exemplary individuals; identify a second set of individuals of the plurality of individuals that are similar to at least one individual of the first set wherein the similarity is based on features associated with the at least one individual and the second set of individuals; label the second set of individuals as exemplary. ### Claim 15 (depends on 10) The electronic device of claim 10, wherein the influence of the weighted nodes on the unweighted nodes is determined using Poisson's equation or Laplace's equation. ### Claim 16 A method performed by one or more processors and comprising: obtaining, from one or more data sources, one or more data sets associated with a plurality of individuals; determining features associated with the plurality of individuals; determining similarities among the plurality of individuals, based on the determined features; generating data representing a graph having a plurality of nodes, wherein the nodes of the graph are associated with the plurality of individuals, edges between the nodes represent the determined similarities, wherein the nodes include weighted nodes that are associated with a first subset of the plurality of individuals and unweighted nodes; applying a weight to the unweighted nodes of the graph, wherein the weights are based on an influence exerted by the weighted nodes on the unweighted nodes; providing output associated with the graph and the applied weights. ### Claim 17 (depends on 16) The method of claim 16, wherein the edges of the graph are weighted and the weight of an edge is based on the determined similarity between individuals associated with the nodes connected by the edge. ### Claim 18 (depends on 16) The method of claim 16, wherein one or more individuals of the first subset are labeled as positive and one or more individuals of the first subset are labeled as negative. ### Claim 19 (depends on 16) The method of claim 16, further comprising: labeling one or more individuals of the first subset as exemplary comprising: labeling a first set of individuals of the plurality of individuals as exemplary, wherein the first set of individuals includes at least one individual that is associated with a subset of features that is correlated to qualities associated with exemplary individuals; identifying a second set of individuals of the plurality of individuals that are similar to at least one individual of the first set, wherein the similarity is based on one or more features associated with the at least one individual and the second set of individuals; and labeling the second set of individuals as exemplary. ### Claim 20 (depends on 16) The method of claim 16, wherein the influence exerted by the weighted nodes on the unweighted nodes is determined by assigning the weighted nodes point charges and calculating an electrical potential of the unweighted nodes.