# US 12,602,661 B2 · System for searching and correlating online activity with individual classification factors > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,602,661 B2 - Document ID: US12602661B2 - Status: Granted - Kind: B2 - Application number: 15/365,209 - Filed: 2016-11-30 - Published: 2017-06-01 - Granted: 2026-04-14 - Earliest claimed: 2015-11-30 - Assignees: FAMA Technologies, Inc. - Applicants: FAMA Technologies, Inc. - Inventors: Benjamin Yale Mones; Collin James Scangarella; Alec Serge Andronikov - Web page: https://patentgrep.com/patent/US12602661B2 ## CPC - G06Q 10/1053: Employment or hiring - G06F 16/24578: {using ranking} - G06F 16/2462: {Approximate or statistical queries} - G06F 16/25: Integrating or interfacing systems involving database management systems - G06F 16/258: {Data format conversion from or to a database} - G06F 16/285: {Clustering or classification} - G06N 20/00: Machine learning - G06N 5/025: Extracting rules from data - G06Q 10/40: Business processes related to social networking or social networking services - G06Q 10/42: Determination of affinities or common interests between users - G06Q 10/48: using social graphs - 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 - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract An attribute associated with an individual being screened for a category of conduct is obtained. A first set of data associated with the individual is obtained, based at least in part on the attribute, with a form of the first set of data being one of a plurality of disparate forms. A second set of data associated with the individual is obtained, based at least in part on the attribute, with a form of the second set of data being different from the form of the first set of data. The first set of data and the second set of data are transformed into a standard data structure. Contents of the standard data structure are input to an algorithm. A result from the algorithm is received that indicates that the individual is associated with the category of conduct, and the result is output to a user interface. ## Claims ### Claim 1 A computer-implemented method, comprising: obtaining a relevant data preference indicating a category of data that is relevant for a search result; extracting a key attribute from initial source data corresponding to an individual, the key attribute being personally identifiable information usable to identify additional information that is associated with the individual; obtaining, based at least in part on the key attribute, the additional information from an online source associated with the individual, wherein the additional information includes at least one digital image; transforming, at least in part by performing object recognition on the at least one digital image, the additional information that includes the at least one digital image into a first structured dataset comprising a plurality of fields, the first structured dataset selected from a plurality of structured data sets based on at least a type of the online source, wherein individual structured datasets of the plurality of structured datasets comprise at least one first field of the plurality of fields that identifies the type of the online data source and at least one second field of the plurality of fields that contains data specific to the type of the online data source of the individual structured dataset; determining to an amount of statistical certainty, at least in part by passing the first structured dataset as input to a machine learning algorithm, that the additional information is associated with the category of data, the machine learning algorithm having been previously trained with prior inter-rater agreements on the category of data; generating the search result indicating, based at least in part on the amount of statistical certainty, that the at least one digital image is associated with the category of data; and providing the search result in a user interface. ### Claim 2 (depends on 1) The computer-implemented method of claim 1, wherein obtaining the additional information includes: calling, using the key attribute as a parameter value, an application programming interface to the online source; and receiving, as a response, the additional information. ### Claim 3 (depends on 1) The computer-implemented method of claim 1, wherein: the category is associated with a weight that indicates a level of importance of the category; and the search result is further generated based at least in part on the weight associated with the category. ### Claim 4 (depends on 1) The computer-implemented method of claim 1, wherein obtaining the relevant data preference includes receiving, via selection of the category of data from a user interface, a request to identify online activity associated with the individual that corresponds to the category of data. ### Claim 5 A system, comprising: one or more processors; and memory including instructions that, as a result of execution by the one or more processors, cause the system to: receive a selection of a screening priority for screening an individual; extract an attribute from source data that is associated with the individual; obtain, based at least in part on the attribute, unstructured data associated with the individual from an online source, the unstructured data including a digital image; transform the unstructured data including the digital image into a structured dataset, the structured dataset selected from a plurality of structured data sets based on a type of the online source, the structured dataset comprising a data field that corresponds to the type of the online data source; make a determination, at least in part by passing the structured dataset to a machine learning algorithm, within an amount of statistical certainty that the digital image is associated with the screening priority; and output the determination to a user interface. ### Claim 6 (depends on 5) The system of claim 5, wherein: the attribute is associated with a weight; and the instructions further include instructions that cause the system to: calculate, based at least in part on the weight, a probability that the unstructured data is associated with the individual; and determine, based at least in part on the probability calculated being a value relative to a threshold, that the individual is associated with the unstructured data. ### Claim 7 (depends on 5) The system of claim 5, wherein: the system is provided to a user as a job applicant screening service; and the instructions are executed in response to the user submitting a request to screen the individual as a prospective job applicant. ### Claim 8 (depends on 5) The system of claim 5, wherein: the source data is a document that presents details on a background of the individual; and the instructions further include instructions that cause the system to parse the document to identify the attribute; and extract the attribute from the document. ### Claim 9 (depends on 5) The system of claim 5, wherein the machine learning algorithm is trained from a set of training data, the set of training data obtained by collecting inter-rater agreements between a plurality of individuals judging particular target data. ### Claim 10 (depends on 5) The system of claim 5, wherein the instructions further cause the system to: store the structured dataset in persistent storage with a plurality of structured datasets; and retrain the machine learning algorithm at least in part from the plurality of structured datasets. ### Claim 11 (depends on 5) The system of claim 5, wherein the instructions that cause the system to transform the unstructured data include instructions that cause the system to: obtain, by passing at least a portion of the unstructured data as input to a visual recognition application, one or more text labels that describe contents of the unstructured data; and generate the structured dataset at least in part from the one or more text labels. ### Claim 12 (depends on 5) The system of claim 5, wherein the unstructured data further includes one or more of: a check-in, digital audio data, or digital video data. ### Claim 13 A non-transitory computer-readable storage medium having stored thereon executable instructions that, as a result of being executed by one or more processors of a computer system, cause the computer system to at least: train, based on inter-rater agreements on whether historical data corresponds to a category of conduct, an algorithm to identify the category of conduct in a plurality of disparate forms of data, the plurality of disparate forms including digital image data; obtain an attribute associated with an individual being screened; obtain, based at least in part on the attribute, a first set of data associated with the individual from an online data source, the first set of data corresponding to the digital image data, wherein the online data source comprises at least one of a social media site, a website, or a search engine; obtain, based at least in part on the attribute, a second set of data associated with the individual, the second set of data corresponding to another of the plurality of disparate forms; transform the first set of data that at least corresponds to the digital image data and the second set of data into a standard data structure determined from a plurality of standard data structures based at least in part on the online data source; input contents of the standard data structure to the algorithm; receive a result from the algorithm that indicates that the first set of data associated with the individual is associated with the category of conduct; and output the result to a user interface. ### Claim 14 (depends on 13) The non-transitory computer-readable storage medium of claim 13, wherein: the computer system is a computer system of a background investigation service; and the executable instructions are executed in response to a request from a user of the background investigation service requesting to investigate a background of the individual. ### Claim 15 (depends on 13) The non-transitory computer-readable storage medium of claim 13, wherein: the computer system is a computer system of a dating service; and the executable instructions are executed in response to a request from a client of the dating service requesting to screen the individual as a potential dating match for the client. ### Claim 16 (depends on 13) The non-transitory computer-readable storage medium of claim 13, wherein the executable instructions further cause the computer system to: obtain source data associated with the individual from a user; and extract the attribute from the source data. ### Claim 17 (depends on 13) The non-transitory computer-readable storage medium of claim 13, wherein the executable instructions further cause the computer system to: provide, for selection by a user, a set of categories; and receive a selection from the user that indicates the category of conduct; and storing the selection in persistent storage. ### Claim 18 (depends on 13) The non-transitory computer-readable storage medium of claim 13, wherein: the algorithm is a machine learning algorithm trained from a dataset developed at least in part through inter-rater agreements of a plurality of individuals; and the machine learning algorithm receives the standard data structure as input and outputs a value that indicates a statistical certainty that the first set of data and the second set of data are associated with the category of conduct. ### Claim 19 (depends on 13) The non-transitory computer-readable storage medium of claim 13, wherein the executable instructions that cause the computer system to transform the first set of data include executable instructions that cause the computer system to: obtain, by passing at least a subset of the first set of data as input to a natural language processing algorithm, one or more text labels that describe contents of the at least a subset; and generate the standard data structure at least in part from the one or more text labels. ### Claim 20 (depends on 13) The non-transitory computer-readable storage medium of claim 13, wherein the executable instructions that cause the computer system to output the result to a user interface include executable instructions that cause the computer system to: generate a report that organizes the first set of data and the second set of data into one or more categories; and in response to receiving a selection of one the one or more categories, display the first set of data and the second set of data in the user interface. ### Claim 21 (depends on 1) The computer-implemented method of claim 1, wherein the first structured dataset further comprises a serializable format, wherein the first structured dataset is selected from the plurality of structured datasets further based on an indication that the structured dataset will be communicated over a network. ### Claim 22 (depends on 1) The computer-implemented method of claim 1, wherein the first structured dataset further comprises a non-serializable format, wherein the first structured dataset is selected from the plurality of structured datasets further based on an indication that the structured dataset will be persisted in database.