Abstract:Abstract A significant component of the online learning platform is the online exercise assessment system, which has access to a wealth of past student exercise data that may be used for data mining research. However, the data from the present online exercise system is not efficiently used, making each exercise less relevant for students and decreasing their interest and interaction with the teacher as she explains the activities. In light of this, this research creates an exercise knowledge map based on the connections between workouts, knowledge points, and previous tournaments. The neural matrix was then improved using cross-feature sharing and feature augmentation units to deconstruct the workout recommendation model. The study also developed an interactive text sentiment analysis model based on the expansion of the self-associative word association network to assess how students interacted after the introduction of the personalized exercise advice teaching approach. The outcomes demonstrated that the suggested model’s mean diversity value at completion was 0.93, an increase of 0.14 and 0.23 over collaborative filtering algorithm and DeepFM (deep factor decompose modle), respectively, and that the proposed model’s final convergence value was 92.3%, an improvement of 2.3 and 4.1% over the latter two models. The extended model used in the study outperformed the support vector machine (SVM) and Random Forest models in terms of accuracy by 5.9 and 1.7%, respectively. In terms of F 1 value indicator, the model proposed by the research has a value of 90.4%, which is 2.5 and 2.1% higher than the SVM model and Random Forest model; in terms of recall rate indicators, the model proposed by the research institute has a value of 94.3%, which is an increase of 6.2 and 9.8% compared to the latter two models. This suggests that the study’s methodology has some application potential and is advantageous in terms of customized recommendation and interactive sentiment recognition.

Wse-MF: A Weighting-based Student Exercise Matrix Factorization Model

Online English writing teaching method that enhances teacher–student interaction

Self-Paced Learning for Matrix Factorization

Exercise Difficulty Prediction in Online Education Systems

Fast Matrix Factorization with Non-Uniform Weights on Missing Data.

Enhancing Cognitive Diagnosis using Un-interacted Exercises: A Collaboration-aware Mixed Sampling Approach

A Multilayered-and-Randomized Latent Factor Model for High-Dimensional and Sparse Matrices

Exercise-Enhanced Sequential Modeling for Student Performance Prediction

Learning Behavior-Aware Cognitive Diagnosis for Online Education Systems

Knowledge Graph Enhanced Intelligent Tutoring System Based on Exercise Representativeness and Informativeness

Cognitive Modelling for Predicting Examinee Performance

Innovative models for enhanced student adaptability and performance in educational environments

Exploiting Non-Interactive Exercises in Cognitive Diagnosis.

Exponentially Convergent Algorithms for Supervised Matrix Factorization

Modeling and Predicting Learning Behavior in MOOCs

A Learned Simulation Environment to Model Student Engagement and Retention in Automated Online Courses

Graph-based Exercise- and Knowledge-Aware Learning Network for Student Performance Prediction

Local Weighted Matrix Factorization for Implicit Feedback Datasets.

Distributed matrix factorization based on fast optimization for implicit feedback recommendation

A Cognitive Diagnosis Framework Based on Peer Assessment.

Pull together: Option-weighting-enhanced mixture-of-experts knowledge tracing