Gamification elements-based engineering students’ learning behavior analysis and performance modelling system using ZSH-FUZZY and Deep-JRGNN

Educators can identify the students who are at risk of underperformance by predicting their performance. However, the existing works didn’t concentrate on specific gamification elements for accurate performance modelling. Therefore, this paper presents gamification elements-based engineering students learning behavior analysis and performance modelling using Deep-JRGNN and ZSH-Fuzzy. Initially, the student’s performance data are taken; then, they are pre-processed. Afterward, binning and hexbin plot construction are performed. Then, the features are extracted from the hexbin plot. Likewise, pre-processed data are clustered by employing WCTQEK-Means. For the clustered data, the temporal behavioral analysis is done by utilizing the DACWT. Next, the temporal features and behavioral features are extracted. Afterward, the correlation analysis is performed. Then, engineering students’ learning behavior is analyzed by using the percentile rank calculation and ZSH-Fuzzy. Likewise, the gamification elements are considered from the questionnaires. For that, percentile calculation is done and classes are identified by using ZSH-Fuzzy. Then, labelling is done by using ZSH-Fuzzy. Finally, the performance modelling is performed by using Deep-JRGNN, in which the LIME-based DeepXplainer provides a deep explanation of the outcomes. The results proved that the proposed model achieved a high accuracy of 98.74%, which outperformed conventional methods.