# US 12,001,643 B1 · Age prediction of end users based on input device data > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,001,643 B1 - Document ID: US12001643B1 - Status: Granted - Kind: B1 - Application number: 18/082,451 - Filed: 2022-12-15 - Published: 2024-06-20 - Granted: 2024-06-04 - Earliest claimed: 2022-12-15 - Assignees: SardineAI Corp. - Applicants: SardineAI Corp. - Inventors: Chandan Garg; Zehao Xu; Jeffrie Joshua Lazarus George; Srikant Rao; Andrew Tausz - Web page: https://patentgrep.com/patent/US12001643B1 ## CPC - G06F 3/04812: Interaction techniques based on cursor appearance or behaviour, e.g. being affected by the presence of displayed objects - G06F 11/3438: {monitoring of user actions (tracking the activity of the user H04L67/535)} - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract The present technology includes receiving event data of a user from a user device while the user interacts with a graphical user interface (GUI) to access a service, where the event data includes behavioral biometrics of the user obtained from one or more input devices of the user device, and predicting, based on the event data, a projected age of the user, where predicting the age of the user includes inputting the event data into a machine learning model, where the machine learning model is configured to receive event data and output an age prediction for a user associated with the event data. ## Claims ### Claim 1 A computer-implemented method comprising: receiving event data of a user from a user device while the user performs an interaction with a graphical user interface (GUI) to access a first service of a first platform, wherein the event data includes behavioral biometrics of the user obtained from one or more input devices of the user device; receiving additional event data of the user from the user device while the user performs a second interaction with the GUI to access a second service of a second platform, wherein the additional event data includes additional behavioral biometrics of the user obtained from the one or more input devices of the user device; and predicting, based on the event data and the additional event data, a projected age of the user, wherein predicting the age of the user includes inputting the event data and the additional event data into a machine learning model, wherein the machine learning model is configured to receive the event data and the additional event data and output an age prediction for the user associated with the event data. ### Claim 2 (depends on 1) The computer-implemented method of claim 1, further comprising: determining, based on the age prediction output from the machine learning model for the event data from the user device as the user interacts with the GUI of the service, whether the interaction is fraudulent. ### Claim 3 (depends on 1) The computer-implemented method of claim 1, further comprising: receiving, by an evaluation service, a request from the service regarding the user; and outputting, by the evaluation service, the projected age to the service. ### Claim 4 (depends on 2) The computer-implemented method of claim 2, wherein determining whether the interaction is fraudulent is further based on whether the event data matches the additional event data. ### Claim 5 (depends on 1) The computer-implemented method of claim 1, further comprising: receiving inputted data from the user device, wherein the inputted data is data that the user input during the interaction, and wherein the inputted data includes an age input for user, and wherein determining whether the interaction is fraudulent is further based on whether the projected age is within a threshold range of the age input for the user. ### Claim 6 (depends on 1) The computer-implemented method of claim 1, wherein the user device is a mobile phone, and wherein the one or more input devices include at least one of a gyroscope and an accelerometer. ### Claim 7 (depends on 1) The computer-implemented method of claim 1, wherein the event data includes types of data including at least one of a timestamp, movement of a mouse, clicking of the mouse, scrolling a scroll wheel of the mouse, a mousepad input, x-y coordinates of a cursor of the mouse, a key input on a keyboard, a key stroke, a keypress down of a key on the keyboard, a keypress release of the key on the keyboard, and field switching. ### Claim 8 (depends on 7) The computer-implemented method of claim 7, further comprising: deriving the behavioral biometrics based on at least one of the types of data. ### Claim 9 (depends on 1) The computer-implemented method of claim 1, wherein the behavioral biometrics include at least one of a manner of switching from one field to another field, velocity of moving the mouse, jitters, typing speed, hesitation percentage, time spent filling out one or more fields of a form, input habits, and a frequency of typographic errors. ### Claim 10 (depends on 1) The computer-implemented method of claim 1, further comprising: training the machine learning model to receive the event data and predict the projected age, the training comprising: providing prediction training inputs to the machine learning model, wherein the prediction training inputs include a training data set of event data set and a training data set of projected ages. ### Claim 11 (depends on 10) The computer-implemented method of claim 10, wherein the event data includes one or more different types of data, and wherein a first type of data has a higher variance than a second type of data and the second type of data has a lower variance than the first type of data, and wherein training the machine learning model further comprises: increasing a weight of the first type of data for predicting the projected age. ### Claim 12 A system comprising: a storage configured to store instructions; a processor configured to execute the instructions and cause the processor to: receive event data of a user from a user device while the user performs an interaction with a graphical user interface (GUI) to access a first service of a first platform, wherein the event data includes behavioral biometrics of the user obtained from one or more input devices of the user device; receive additional event data of the user from the user device while the user performs a second interaction with the GUI to access a second service of a second platform, wherein the additional event data includes additional behavioral biometrics of the user obtained from the one or more input devices of the user device; and predict, based on the event data and the additional event data, a projected age of the user, wherein predicting the age of the user includes inputting the event data and the additional event data into a machine learning model, wherein the machine learning model is configured to receive the event data and the additional event data and output an age prediction for the user associated with the event data. ### Claim 13 (depends on 12) The system of claim 12, wherein the instructions further cause the processor to: determine, based on the age prediction output from the machine learning model for the event data from the user device as the user interacts with the GUI of the service, whether the interaction is fraudulent. ### Claim 14 (depends on 12) The system of claim 12, wherein the instructions further cause the processor to: receive, by an evaluation service, a request from the service regarding the user; and output, by the evaluation service, the projected age to the service. ### Claim 15 (depends on 14) The system of claim 14, wherein determining whether the interaction is fraudulent is further based on whether the event data matches the additional event data. ### Claim 16 (depends on 12) The system of claim 12, wherein the instructions further cause the processor to: receive inputted data from the user device, wherein the inputted data is data that the user input during the interaction, and wherein the inputted data includes an age input for user, and wherein determining whether the interaction is fraudulent is further based on whether the projected age is within a threshold range of the age input for the user. ### Claim 17 A non-transitory computer readable medium comprising instructions, which when executed by a computing system, cause the computing system to: receive event data of a user from a user device while the user performs an interaction with a graphical user interface (GUI) to access a first service of a first platform, wherein the event data includes behavioral biometrics of the user obtained from one or more input devices of the user device; receive additional event data of the user from the user device while the user performs a second interaction with the GUI to access a second service of a second platform, wherein the additional event data includes additional behavioral biometrics of the user obtained from the one or more input devices of the user device; and predict, based on the event data and the additional event data, a projected age of the user, wherein predicting the age of the user includes inputting the event data and the additional event data into a machine learning model, wherein the machine learning model is configured to receive the event data and the additional event data and output an age prediction for the user associated with the event data. ### Claim 18 (depends on 17) The non-transitory computer readable medium of claim 17, wherein the event data includes types of data including at least one of a timestamp, movement of a mouse, clicking of the mouse, scroll a scroll wheel of the mouse, a mousepad input, x-y coordinates of a cursor of the mouse, a key input on a keyboard, a key stroke, a keypress down of a key on the keyboard, a keypress release of the key on the keyboard, and field switching. ### Claim 19 (depends on 17) The non-transitory computer readable medium of claim 17, wherein the instructions, when executed by the computing system, further cause the computing system to: train the machine learning model to receive the event data and predict the projected age, the training comprising: providing prediction training inputs to the machine learning model, wherein the prediction training inputs include a training data set of event data set and a training data set of projected ages. ### Claim 20 (depends on 19) The non-transitory computer readable medium of claim 19, wherein the event data includes one or more different types of data, and wherein a first type of data has a higher variance than a second type of data and the second type of data has a lower variance than the first type of data, and wherein training the machine learning model further comprises: increasing a weight of the first type of data for predicting the projected age.