# US 12,182,765 B2 · Automated credential processing system > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,182,765 B2 - Document ID: US12182765B2 - Status: Granted - Kind: B2 - Application number: 18/122,333 - Filed: 2023-03-16 - Published: 2023-09-21 - Granted: 2024-12-31 - Earliest claimed: 2022-03-17 - Assignees: Liveperson, Inc. - Applicants: Liveperson, Inc. - Inventors: Dev Sharma; Marco Ambrosio; Sam Kirby - Web page: https://patentgrep.com/patent/US12182765B2 ## CPC - G06Q 10/1053: Employment or hiring - G06F 16/24578: {using ranking} - G06F 16/285: {Clustering or classification} - G06N 20/00: Machine learning - G06Q 10/063112: {Skill-based matching of a person or a group to a task} - G06N 5/022: Knowledge engineering; Knowledge acquisition - 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 Disclosed embodiments provide a framework for processing sets of credentials in real-time using machine learning models to identify requisitions that can be recommended to users. In response to receiving a set of credentials from a user, a system assigns a classification to the set of credentials based on a set of characteristics associated with the set of credentials. A machine learning model is used to assign tags to the text of the set of credentials. These tags correspond to the set of characteristics. Using these tags, a set of open requisitions are identified and provided to the user. ## Claims ### Claim 1 A computer-implemented method comprising: receiving a set of credentials associated with a user, wherein the set of credentials identifies a set of characteristics associated with the user; applying a category identification model to the set of credentials to generate a requisition category associated with the user, wherein the requisition category identifies a domain for the set of credentials, and wherein the category identification model is trained using previously processed sets of credentials and their respective requisition categories; applying a skill recognition model to the requisition category and the set of credentials after the category identification model generates the requisition category, wherein the skill recognition model generates a plurality of candidate tags associated with the domain identified by the requisition category, wherein the plurality of candidate tags identify skills associated with the set of characteristics, and wherein the skill recognition model is trained using the previously processed sets of credentials and their respective tags; removing a set of candidate tags from the plurality of candidate tags after the skill recognition model generates the plurality of candidate tags, wherein the set of candidate tags include personal identifiable information associated with the user, and wherein the set of candidate tags are removed to generate one or more anonymized tags for the set of credentials; identifying one or more open requisitions using the one or more anonymized tags; ranking the one or more open requisitions according to confidence scores associated with the one or more anonymized tags; receiving feedback associated with a third-party system, wherein the feedback includes an indication of whether the user qualifies for the one or more open requisitions; updating the category identification model and the skill recognition model using the one or more open requisitions and the feedback; and updating the skill recognition model using the one or more anonymized tags, wherein updating the skill recognition model reduces bias associated with determining whether the user qualifies for the one or more open requisitions. ### Claim 2 (depends on 1) The computer-implemented method of claim 1, wherein the skill recognition model processes the set of credentials using named entity recognition to generate the plurality of candidate tags. ### Claim 3 (depends on 1) The computer-implemented method of claim 1, wherein identifying the one or more open requisitions includes matching the one or more anonymized tags to a set of requisitions to identify the one or more open requisitions. ### Claim 4 (depends on 1) The computer-implemented method of claim 1, wherein the one or more anonymized tags are associated with one or more skills, and wherein the one or more skills correspond to the one or more open requisitions. ### Claim 5 (depends on 1) The computer-implemented method of claim 1, further comprising: obtaining a set of requisitions; and assigning requisition tags to the set of requisitions, wherein the one or more anonymized tags are matched to the requisition tags to identify the one or more open requisitions. ### Claim 6 (depends on 1) The computer-implemented method of claim 1, wherein the one or more anonymized tags are generated to determine whether the user qualifies for requisitions associated with the third-party system. ### Claim 7 (depends on 1) The computer-implemented method of claim 1, wherein the category identification model and the skill recognition model are updated to increase a probability that the user qualifies for requisitions associated with the third-party system. ### Claim 8 (depends on 1) The computer-implemented method of claim 1, further comprising: determining confidence scores for the one or more anonymized tags, wherein the confidence scores are determined by processing the set of credentials using the skill recognition model, wherein the confidence scores indicate an extent to which the one or more anonymized tags accurately describe the set of characteristics, and wherein the one or more open requisitions are identified further based on the confidence scores. ### Claim 9 (depends on 1) The computer-implemented method of claim 1, further comprising: determining a set of weights for the one or more anonymized tags, wherein the set of weights are determined based on a weight schema generated by the third-party system, and wherein the one or more open requisitions are identified further based on the set of weights. ### Claim 10 A system, comprising: one or more processors; and memory storing thereon instructions that, as a result of being executed by the one or more processors, cause the system to: receive a set of credentials associated with a user, wherein the set of credentials identifies a set of characteristics associated with the user; apply a category identification model to the set of credentials to generate a requisition category associated with the user, wherein the requisition category identifies a domain for the set of credentials, and wherein the category identification model is trained using previously processed sets of credentials and their respective requisition categories; apply a skill recognition model to the requisition category and the set of credentials after the category identification model generates the requisition category, wherein the skill recognition model generates a plurality of candidate tags associated with the domain identified by the requisition category, wherein the plurality of candidate tags identify skills associated with the set of characteristics, and wherein the skill recognition model is trained using the previously processed sets of credentials and their respective tags; remove a set of candidate tags from the plurality of candidate tags after the skill recognition model generates the plurality of candidate tags, wherein the set of candidate tags include personal identifiable information associated with the user, and wherein the set of candidate tags are removed to generate one or more anonymized tags for the set of credentials; identify one or more open requisitions using the one or more anonymized tags; rank the one or more open requisitions according to confidence scores associated with the one or more anonymized tags; receive feedback associated with a third-party system, wherein the feedback includes an indication of whether the user qualifies for the one or more open requisitions; update the category identification model and the skill recognition model using the one or more open requisitions and the feedback; and update the skill recognition model using the one or more anonymized tags, wherein updating the skill recognition model reduces bias associated with determining whether the user qualifies for the one or more open requisitions. ### Claim 11 (depends on 10) The system of claim 10, wherein the skill recognition model processes the set of credentials using named entity recognition to generate the plurality of candidate tags. ### Claim 12 (depends on 10) The system of claim 10, wherein identifying the one or more open requisitions includes matching the one or more anonymized tags to a set of requisitions to identify the one or more open requisitions. ### Claim 13 (depends on 10) The system of claim 10, wherein the one or more anonymized tags are associated with one or more skills, and wherein the one or more skills correspond to the one or more open requisitions. ### Claim 14 (depends on 10) The system of claim 10, wherein the instructions further cause the system to: obtain a set of requisitions; and assign requisition tags to the set of requisitions, wherein the one or more anonymized tags are matched to the requisition tags to identify the one or more open requisitions. ### Claim 15 (depends on 10) The system of claim 10, wherein the category identification model and the skill recognition model are updated to increase a probability that the user qualifies for requisitions associated with the third-party system. ### Claim 16 (depends on 10) The system of claim 10, wherein the instructions further cause the system to: determine confidence scores for the one or more anonymized tags, wherein the confidence scores are determined by processing the set of credentials using the skill recognition model, wherein the confidence scores indicate an extent to which the one or more anonymized tags accurately describe the set of characteristics, and wherein the one or more open requisitions are identified further based on the confidence scores. ### Claim 17 (depends on 10) The system of claim 10, wherein the instructions further cause the system to: determine a set of weights for the one or more anonymized tags, wherein the set of weights are determined based on a weight schema generated by the third-party system, and wherein the one or more open requisitions are identified further based on the set of weights. ### Claim 18 A non-transitory, computer-readable storage medium storing thereon executable instructions that, as a result of being executed by one or more processors of a computer system, cause the computer system to: receive a set of credentials associated with a user, wherein the set of credentials identifies a set of characteristics associated with the user; apply a category identification model to the set of credentials to generate a requisition category associated with the user, wherein the requisition category identifies a domain for the set of credentials, and wherein the category identification model is trained using previously processed sets of credentials and their respective requisition categories; apply a skill recognition model to the requisition category and the set of credentials after the category identification model generates the requisition category, wherein the skill recognition model generates a plurality of candidate tags associated with the domain identified by the requisition category, wherein the plurality of candidate tags identify skills associated with the set of characteristics, and wherein the skill recognition model is trained using the previously processed sets of credentials and their respective tags; remove a set of candidate tags from the plurality of candidate tags after the skill recognition model generates the plurality of candidate tags, wherein the set of candidate tags include personal identifiable information associated with the user, and wherein the set of candidate tags are removed to generate one or more anonymized tags for the set of credentials; identify one or more open requisitions using the one or more anonymized tags; rank the one or more open requisitions according to confidence scores associated with the one or more anonymized tags; receive feedback associated with a third-party system, wherein the feedback includes an indication of whether the user qualifies for the one or more open requisitions; update the category identification model and the skill recognition model using the one or more open requisitions and the feedback; and update the skill recognition model using the one or more anonymized tags, wherein updating the skill recognition model reduces bias associated with determining whether the user qualifies for the one or more open requisitions. ### Claim 19 (depends on 18) The non-transitory, computer-readable storage medium of claim 18, wherein the skill recognition model processes the set of credentials using named entity recognition to generate the plurality of candidate tags. ### Claim 20 (depends on 18) The non-transitory, computer-readable storage medium of claim 18, wherein the one or more open requisitions includes matching the one or more anonymized tags to a set of requisitions to identify the one or more open requisitions. ### Claim 21 (depends on 18) The non-transitory, computer-readable storage medium of claim 18, wherein the one or more anonymized tags are associated with one or more skills, and wherein the one or more skills correspond to the one or more open requisitions. ### Claim 22 (depends on 18) The non-transitory, computer-readable storage medium of claim 18, wherein the executable instructions further cause the computer system to: obtain a set of requisitions; and assign requisition tags to the set of requisitions, wherein the one or more anonymized tags are matched to the requisition tags to identify the one or more open requisitions. ### Claim 23 (depends on 18) The non-transitory, computer-readable storage medium of claim 18, wherein the executable instructions further cause the computer system to: determine confidence scores for the one or more anonymized tags, wherein the confidence scores are determined by processing the set of credentials using the skill recognition model, wherein the confidence scores indicate an extent to which the one or more anonymized tags accurately describe the set of characteristics, and wherein the one or more open requisitions are identified further based on the confidence scores. ### Claim 24 (depends on 18) The non-transitory, computer-readable storage medium of claim 18, wherein the executable instructions further cause the computer system to: determine a set of weights for the one or more anonymized tags, wherein the set of weights are determined based on a weight schema generated by the third-party system, and wherein the one or more open requisitions are identified further based on the set of weights.