Artificial intelligence-driven avatar-based personalized learning techniques

Patent metadata

Field	Value
Publication number	US-12288480-B2
Application number	US-202217580960-A
Country	US
Kind code	B2
Filing date	Jan 21, 2022
Priority date	Jan 21, 2022
Publication date	Apr 29, 2025
Grant date	Apr 29, 2025

Abstract

Methods, apparatus, and processor-readable storage media for artificial intelligence-driven avatar-based personalized learning techniques are provided herein. An example computer-implemented method includes obtaining multiple forms of input data from user devices associated with a user in a virtual learning environment; determining status information for user variables by processing at least a portion of the multiple forms of input data using a first set of artificial intelligence techniques; determining instruction-related modifications for the user by processing, using a second set of artificial intelligence techniques, at least a portion of the multiple forms of input data and at least a portion of the determined status information; implementing, based on the determined instruction-related modifications, modifications to an instructor avatar with respect to the user in the virtual learning environment; and performing one or more automated actions based on user response to the implemented modifications to the instructor avatar.

First claim

What is claimed is:

1. A computer-implemented method comprising:
obtaining multiple forms of input data from one or more user devices associated with a plurality of users in at least one virtual learning environment;
determining status information for one or more user variables by processing at least a portion of the multiple forms of input data using a first set of one or more artificial intelligence techniques, wherein determining status information for one or more user variables comprises detecting at least one of one or more facial expressions of each of the plurality of users and one or more body movements of each of the plurality of users by processing at least a portion of the input data using one or more computer vision algorithms, comprising performing, for each of the plurality of users, a first object detection operation for detecting one or more portions of a human face attributed to the user, and performing, for each of the plurality of users, a second object detection operation, different from the first object detection operation, for detecting one or more other portions of a human body attributed to the user, using one or more artificial intelligence-related libraries associated with object detection, filtering out at least a portion of the input data unrelated to the object detection, and applying at least one object feature detection algorithm to the filtered input data;
determining one or more instruction-related modifications for each of the plurality of users by processing, using a second set of one or more artificial intelligence techniques, at least a portion of the multiple forms of input data and at least a portion of the determined status information, wherein determining one or more instruction-related modifications for each of the plurality of users comprises dynamically generating new content related to at least one area of instruction identified as requiring additional time for additional instruction for the respective user, and wherein dynamically generating new content comprises processing data in at least one knowledge graph, related to the at least one virtual learning environment, to generate and answer natural language questions, wherein the at least one knowledge graph comprises knowledge graph embeddings representing multiple predicates as vectors of one or more designated dimensions;
implementing, based at least in part on the one or more determined instruction-related modifications, one or more modifications to at least one instructor avatar with respect to each of the plurality of users in the at least one virtual learning environment, wherein implementing one or more modifications to at least one instructor avatar comprises configuring communication to each of the plurality of users through the at least one instructor avatar in a language preferred by each of the plurality of users; and
performing one or more automated actions based at least in part on user response to the one or more implemented modifications to the at least one instructor avatar;
wherein the method is performed by at least one processing device comprising a processor coupled to a memory.

2. The computer-implemented method of claim 1 , wherein determining status information for one or more user variables comprises determining at least one emotional status of each of the plurality of users based at least in part on the one or more detected facial expressions.

3. The computer-implemented method of claim 1 , wherein determining status information for one or more user variables comprises determining a level of user engagement with instruction in the at least one virtual learning environment based at least in part on the one or more detected body movements.

4. The computer-implemented method of claim 1 , wherein implementing one or more modifications to at least one instructor avatar comprises modifying at least one facial expression exhibited by the at least one instructor avatar to each of the plurality of users.

5. The computer-implemented method of claim 1 , wherein implementing one or more modifications to at least one instructor avatar comprises modifying a tone of communication output from the at least one instructor avatar to each of the plurality of users.

6. The computer-implemented method of claim 1 , wherein performing one or more automated actions comprises automatically training, using at least a portion of the user response, at least one of the first set of one or more artificial intelligence techniques and the second set of one or more artificial intelligence techniques.

7. The computer-implemented method of claim 1 , wherein determining one or more instruction-related modifications for each of the plurality of users comprises determining one or more learning patterns attributed to each of the plurality of users by processing user data related to one or more of historical performance, user background information, one or more user preferences, information pertaining to time spent learning by each of the plurality of users, and information pertaining to repeat activity within the at least one virtual learning environment by each of the plurality of users.

8. The computer-implemented method of claim 1 , wherein obtaining multiple forms of input data from one or more user devices associated with a plurality of users in at least one virtual learning environment comprises obtaining multiple forms of input data from one or more of at least one camera associated with each of the plurality of users, at least one Internet of Things sensors associated with each of the plurality of users, and at least one wearable device worn by each of the plurality of users.

9. The computer-implemented method of claim 1 , wherein the first set of one or more artificial intelligence techniques comprises the same one or more artificial intelligence techniques as the second set of one or more artificial intelligence techniques.

10. The computer-implemented method of claim 1 , wherein the first set of one or more artificial intelligence techniques comprises a distinct set of one or more artificial intelligence techniques from the second set of one or more artificial intelligence techniques.

11. A non-transitory processor-readable storage medium having stored therein program code of one or more software programs, wherein the program code when executed by at least one processing device causes the at least one processing device:
to obtain multiple forms of input data from one or more user devices associated with a plurality of users in at least one virtual learning environment;
to determine status information for one or more user variables by processing at least a portion of the multiple forms of input data using a first set of one or more artificial intelligence techniques, wherein determining status information for one or more user variables comprises detecting at least one of one or more facial expressions of each of the plurality of users and one or more body movements of each of the plurality of users by processing at least a portion of the input data using one or more computer vision algorithms, comprising performing, for each of the plurality of users, a first object detection operation for detecting one or more portions of a human face attributed to the user, and performing, for each of the plurality of users, a second object detection operation, different from the first object detection operation, for detecting one or more other portions of a human body attributed to the user, using one or more artificial intelligence-related libraries associated with object detection, filtering out at least a portion of the input data unrelated to the object detection, and applying at least one object feature detection algorithm to the

Assignees

• Dell Products Lp

Inventors

• Sha Danqing
• Wang Zijia
• Bruno Eric
• Seibel Amy
• Jia Zhen

Classifications

• G06V40/174

  Facial expression recognition · CPC title

• G06V40/20

  Movements or behaviour, e.g. gesture recognition (recognition of facial expressions G06V40/16) · CPC title

• G06V40/70

  Multimodal biometrics, e.g. combining information from different biometric modalities · CPC title

• G06N20/00

  Machine learning · CPC title

• G06N3/006

  based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO] · CPC title