US 12,260,638 B1
Adaptive learning systems utilizing machine learning techniques
Jiarui Tong, Highland Park, NJ (US); Guodong Zhao, Pittsburgh, PA (US); Jiachen Gong, Pittsburgh, PA (US); Yuchen Xie, Pittsburgh, PA (US); and Miaojun Xu, Pittsburgh, PA (US)
Assigned to Squirrel Ai Learning Inc., Bellevue, WA (US)
Filed by Yixue Education Inc., Highland Park, NJ (US)
Filed on Dec. 31, 2020, as Appl. No. 17/139,984.
Int. Cl. G06N 5/02 (2023.01); G06F 3/04842 (2022.01); G06V 20/20 (2022.01)
CPC G06V 20/20 (2022.01) [G06F 3/04842 (2013.01); G06N 5/02 (2013.01)] 14 Claims
OG exemplary drawing
 
1. An apparatus, comprising:
a memory and
a processing device communicably coupled to the memory, the processing device to:
receive a first user input providing a portion of an annotation to be associated with a first practice problem and a first user;
identify one or more knowledge components relating to the first user input, wherein the processing device processes the first user input using a trained machine learning model configured to generate one or more outputs indicating the knowledge components included in inputs provided to the trained machine learning model and the knowledge components relating to the inputs;
autocomplete the annotation to include information of one or more of the knowledge components, wherein the information of the one or more of the knowledge components comprises one or more of a definition of a first knowledge component of the knowledge components relating to the first user input and an equation relating to the first knowledge component;
in view of a user request for learning data relating to the first user, present a personalized report comprising information of a first problem set associated with the first user, wherein the first problem set comprises at least one annotated practice problem relating to the first user, wherein, to present the personalized report to the first user, the processing device is further to:
present, in a second user interface, first content corresponding to the at least one annotated practice problem relating to the first user;
present, in the second user interface, second content corresponding to annotation data associated with the at least one annotated practice problem relating to the first user; and
present, on a display, information of a second problem set comprising one or more recommended practice problems, wherein the one or more recommended practice problems are identified based at least in part on predicted tag information relating to a plurality of known practice problems, and wherein the predicted tag information relating to the plurality of known practice problems comprises a plurality of predicted tags associated with a plurality of users and the plurality of known practice problems, wherein the first user input corresponds to a first user interaction with a first user interface element in a first user interface, and wherein, to autocomplete, by the processing device, the annotation to include information relating to one or more of the knowledge components, the processing device is further to present the information relating to the one or more knowledge components in the first user interface element in the first user interface, wherein the annotation data comprises text, images, graphics, video content, and audio content, wherein the one or more knowledge components relating to the annotation data are identified based on the one or more outputs of the trained machine learning model.