一种选项价值增强的多任务知识追踪方法
By employing a multi-task knowledge tracing method that enhances the value of options, and utilizing option weighting and multi-task learning, combined with time series modeling and attention-gated selection, this approach addresses the problem of ignoring the value of interfering options in existing technologies. This enables accurate diagnosis of students' knowledge status and assists in personalized teaching.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- HUAZHONG NORMAL UNIV
- Filing Date
- 2022-10-25
- Publication Date
- 2026-07-17
AI Technical Summary
Existing knowledge tracking technologies ignore the value of distractor options in multiple-choice question assessments, making it difficult to accurately diagnose students' knowledge status and errors, thus affecting the effectiveness of personalized teaching.
We employ a multi-task knowledge tracing method that enhances the value of each option. By calculating the value of each option through an option weighting method, we combine time series modeling and multi-task learning, and use hybrid expert ensemble and attention-gated selection to predict students' future answer performance and specific errors.
It enables accurate diagnosis of students' knowledge status, improving the accuracy of the knowledge tracking model and the auxiliary effect of personalized teaching.
Smart Images
Figure CN115841172B_ABST