Smart input devices with user behavior predictions

Patent metadata

Field	Value
Publication number	US-12554518-B2
Application number	US-202318365508-A
Country	US
Kind code	B2
Filing date	Aug 4, 2023
Priority date	Aug 4, 2023
Publication date	Feb 17, 2026
Grant date	Feb 17, 2026

Abstract

One example method includes detecting an operation of an input device, obtaining a clickstream associated with a user, using information from the clickstream, generating a prediction as to a next action by the user using the input device, and presenting the prediction to the user for possible selection by the user. Selection of the prediction by the user eliminates the need for the user to perform input device manipulations that would otherwise be required if the prediction were not selected.

First claim

What is claimed is:

1 . A method, comprising:
detecting within an application an operation of an input device; obtaining a clickstream associated with the operation of the input device; using metadata from the clickstream and history information of previous clickstreams and corresponding metadata, the clickstream and/or the history information comprising a path, taken by a pointer of the input device, on a display; generating a prediction as to a next action of the application to be performed by the input device; and presenting the prediction to a user for possible selection by the user, wherein the metadata of the clickstream comprises any one or more of an identity of the input device, identification of the application used by the user, and clickstream coordinates corresponding to one or more movements of a pointer associated with the input device, and wherein the prediction comprises acceptance information informing how to accept the prediction without moving the pointer to another coordinate.

2 . The method as recited in claim 1 , wherein the prediction is generated using an autoencoder of a machine learning model.

3 . The method as recited in claim 1 , wherein the information from the clickstream is obtained in real time as the user is manipulating the input device.

4 . The method as recited in claim 1 , wherein the information from the clickstream is obtained from a metadata repository.

5 . The method as recited in claim 1 , wherein the information from the clickstream comprises historical information associated with the user.

6 . The method as recited in claim 1 , wherein after the user selects the prediction, one or more input device selections associated with that prediction are automatically made without requiring further action by the user.

7 . The method as recited in claim 1 , wherein the prediction is generated by a machine learning model that was trained using historical information comprising a history of input device operations performed by the user.

8 . The method as recited in claim 1 , wherein the prediction is generated based on a particular sequence in which events in the information from the clickstream occurred.

9 . The method as recited in claim 1 , wherein the input device comprises any device that can provide the pointer and enable the user to interact with any type of computing system by using the pointer.

10 . A non-transitory storage medium having stored therein instructions that are executable by one or more hardware processors to perform operations comprising:
detecting within an application an operation of an input device; obtaining a clickstream associated with the operation of the input device; using metadata from the clickstream and history information of previous clickstreams and corresponding metadata, the clickstream and/or the history information comprising a path, taken by a pointer of the input device, on a display; generating a prediction as to a next action of the application to be performed by the input device; and presenting the prediction to a user for possible selection by the user, wherein the metadata of the clickstream comprises any one or more of an identity of the input device, identification of the application used by the user, and clickstream coordinates corresponding to one or more movements of a pointer associated with the input device, and wherein the prediction comprises acceptance information informing how to accept the prediction without moving the pointer to another coordinate.

11 . The non-transitory storage medium as recited in claim 10 , wherein the prediction is generated using an autoencoder of a machine learning model.

12 . The non-transitory storage medium as recited in claim 10 , wherein the information from the clickstream is obtained in real time as the user is manipulating the input device.

13 . The non-transitory storage medium as recited in claim 10 , wherein the information from the clickstream is obtained from a metadata repository.

14 . The non-transitory storage medium as recited in claim 10 , wherein the information from the clickstream comprises historical information associated with the user.

15 . The non-transitory storage medium as recited in claim 10 , wherein after the user selects the prediction, one or more input device selections associated with that prediction are automatically made without requiring further action by the user.

16 . The non-transitory storage medium as recited in claim 10 , wherein the prediction is generated by a machine learning model that was trained using historical information comprising a history of input device operations performed by the user.

17 . The non-transitory storage medium as recited in claim 10 , wherein the prediction is generated based on a particular sequence in which events in the information from the clickstream occurred.

18 . The non-transitory storage medium as recited in claim 10 , wherein the input device comprises any device that can provide the pointer and enable the user to interact with any type of computing system by using the pointer.

Assignees

• Dell Products Lp

Inventors

• Popat Rajiv
• Mohanty Bijan Kumar
• Dinh Hung

Classifications

• G06F9/453 Primary

  Help systems · CPC title