# US 12,770,198 B2 · System and method for serverless modification and execution of machine learning algorithms > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,770,198 B2 - Document ID: US12770198B2 - Status: Granted - Kind: B2 - Application number: 18/449,679 - Filed: 2023-08-14 - Published: 2023-11-30 - Granted: 2026-10-20 - Earliest claimed: 2021-03-18 - Assignees: ADP, Inc. - Applicants: ADP, Inc. - Inventors: Roberto Coutinho; Roberto Silveira; Guilherme Gomes - Web page: https://patentgrep.com/patent/US12770198B2 ## CPC - G06Q 40/125: {Finance or payroll} - G06N 3/04: Architecture, e.g. interconnection topology - G06N 3/08: Learning methods - G06Q 10/067: Enterprise or organisation modelling - 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 - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract Systems, methods, and computer-readable storage media for identifying and generating insights using machine learning and serverless computing systems. When events are detected, data from those events is collected and formatted into predefined formats. That formatted data is then used within an instance, where a machine learning algorithm is executed by a serverless computing system using the data within the instance as input. The result from the machine learning algorithm is an insight into the event and event data, which can be presented to users for interpretation. ## Claims ### Claim 1 A method comprising: formatting, by one or more processors coupled with memory, event data from at least one network source; generating, by the one or more processors, a computing instance by inputting the formatted event data into a serverless computing system, the computing instance configured to cause the serverless computing system to apply at least one machine learning algorithm of a plurality of machine learning algorithms to the formatted event data; selecting, by the one or more processors, the at least one machine learning algorithm from the plurality of machine learning algorithms from a data repository, the at least one machine learning algorithm selected based on a type or format of the event data; inserting, by the one or more processors, additional data associated with the event data into the computing instance, wherein insertion of the additional data causes the one or more processors to generate a modified computing instance associated with the event data, the modified computing instance configured to cause the serverless computing system to generate a modified machine learning algorithm by modifying execution of the at least one machine learning algorithm based on the additional data; performing, at runtime, by the one or more processors, at least one of (i) compressing the modified machine learning algorithm responsive to determining the modified machine learning algorithm exceeds a memory or computing resource restriction of the modified computing instance, or (ii) maintaining the modified machine learning algorithm responsive to determining the serverless computing system comprises a feature to automatically scale memory or computing resources; transmitting, by the one or more processors, a request to cause the serverless computing system to execute the modified computing instance using the modified machine learning algorithm; receiving, by the one or more processors, outlier data associated with the event data based on execution of the modified computing instance; and transmitting, by the one or more processors, to a client device, data to cause the client device to display a graphical presentation of the outlier data. ### Claim 2 (depends on 1) The method of claim 1, wherein the event data is formatted into a predefined format, and the method further comprising: receiving, by the one or more processors, a notification regarding an occurrence of the event data from the at least one network source. ### Claim 3 (depends on 1) The method of claim 1, wherein the outlier data corresponds with an insight regarding the event data, and the method comprising: receiving, by the one or more processors, from the serverless computing system, the insight regarding the event data; and displaying, by the one or more processors, the insight regarding the event data. ### Claim 4 (depends on 1) The method of claim 1, wherein the serverless computing system comprises at least one on-demand or cloud-based computing system. ### Claim 5 (depends on 1) The method of claim 1, wherein transmitting the request causes the serverless computing system to select the at least one machine learning algorithm. ### Claim 6 (depends on 1) The method of claim 1, wherein transmitting the request causes the serverless computing system to select a set of business rules from a plurality of business rules. ### Claim 7 (depends on 1) The method of claim 1, wherein transmitting the request causes the serverless computing system to modify execution of the at least one machine learning algorithm based on a comparison of a selected set of business rules and the additional data. ### Claim 8 (depends on 1) The method of claim 1, wherein the outlier data corresponds with an insight regarding the event data, wherein transmitting the request causes the serverless computing system to execute the modified machine learning algorithm using the additional data to generate a modified insight regarding the event data. ### Claim 9 (depends on 1) The method of claim 1, wherein the event data comprises social media network data, and wherein the at least one network source comprises a social media network. ### Claim 10 (depends on 1) The method of claim 1, further comprising: executing, by the one or more processors, a reinforcement learning algorithm on the outlier data; and updating, by the one or more processors, the display of the outlier data based on execution of the reinforcement learning algorithm. ### Claim 11 A system comprising: one or more processors, coupled with memory, to: format event data from at least one network source; generate a computing instance by inputting the formatted event data into a serverless computing system, the computing instance configured to cause the serverless computing system to apply at least one machine learning algorithm of a plurality of machine learning algorithms to the formatted event data; select the at least one machine learning algorithm from the plurality of machine learning algorithms from a data repository, the at least one machine learning algorithm selected based on a type or format of the event data; insert additional data associated with the event data into the computing instance, wherein insertion of the additional data causes the one or more processors to generate a modified computing instance associated with the event data, the modified computing instance configured to cause the serverless computing system to generate a modified machine learning algorithm by modifying execution of the at least one machine learning algorithm based on the additional data; perform, at runtime, at least one of (i) compressing the modified machine learning algorithm responsive to determining the modified machine learning algorithm exceeds a memory or computing resource restriction of the modified computing instance, or (ii) maintaining the modified machine learning algorithm responsive to determining the serverless computing system comprises a feature to automatically scale memory or computing resources; transmit a request to cause the serverless computing system to execute the modified computing instance using the modified machine learning algorithm; receive outlier data associated with the event data based on execution of the modified computing instance; and transmit, to a client device, data to cause the client device to display a graphical presentation of the outlier data. ### Claim 12 (depends on 11) The system of claim 11, wherein the event data is formatted into a predefined format, and the one or more processors further to: receive a notification regarding an occurrence of the event data from the at least one network source. ### Claim 13 (depends on 11) The system of claim 11, wherein the outlier data corresponds with an insight regarding the event data, and the one or more processors further to: receive, from the serverless computing system, the insight regarding the event data; and display the insight regarding the event data. ### Claim 14 (depends on 11) The system of claim 11, wherein the serverless computing system comprises at least one on-demand or cloud-based computing system. ### Claim 15 (depends on 11) The system of claim 11, wherein transmission of the request causes the serverless computing system to select the at least one machine learning algorithm. ### Claim 16 (depends on 11) The system of claim 11, wherein transmission of the request causes the serverless computing system to select a set of business rules from a plurality of business rules. ### Claim 17 (depends on 11) The system of claim 11, wherein transmission of request causes the serverless computing system to modify execution of the at least one machine learning algorithm based on a comparison of a selected set of business rules and the additional data. ### Claim 18 (depends on 11) The system of claim 11, wherein the outlier data corresponds with an insight regarding the event data, wherein transmission of the request causes the serverless computing system to execute the modified machine learning algorithm using the additional data to generate a modified insight regarding the event data. ### Claim 19 (depends on 11) The system of claim 11, wherein the event data comprises social media network data, and wherein the at least one network source comprises a social media network. ### Claim 20 A non-transitory computer readable medium (CRM) having one or more instructions stored thereon, the one or more instructions executable by one or more processors to: format event data from at least one network source; generate a computing instance by inputting the formatted event data into a serverless computing system, the computing instance configured to cause the serverless computing system to apply at least one machine learning algorithm of a plurality of machine learning algorithms to the formatted event data; select the at least one machine learning algorithm from the plurality of machine learning algorithms from a data repository, the at least one machine learning algorithm selected based on a type or format of the event data; insert additional data associated with the event data into the computing instance, wherein insertion of the additional data causes the one or more processors to generate a modified computing instance associated with the event data, the modified computing instance configured to cause the serverless computing system to generate a modified machine learning algorithm by modifying execution of the at least one machine learning algorithm based on the additional data; perform, at runtime, at least one of (i) compressing the modified machine learning algorithm responsive to determining the modified machine learning algorithm exceeds a memory or computing resource restriction of the modified computing instance, or (ii) maintaining the modified machine learning algorithm responsive to determining the serverless computing system comprises a feature to automatically scale memory or computing resources; transmit a request to cause the serverless computing system to execute the modified computing instance using the modified machine learning algorithm; receive outlier data associated with the event data based on execution of the modified computing instance; and transmit, to a client device, data to cause the client device to display a graphical presentation of the outlier data.