Potential Future Directions in Optimization of Students' Performance Prediction System.

Area of Science:

• Educational Psychology
• Data Mining
• Machine Learning

Background:

• Student performance prediction research spans multiple fields, including psychology, data mining, and data analysis.
• Existing prediction approaches often rely heavily on real-world datasets, hindering the integration of psychological and emotional factors.
• A gap exists in synchronizing diverse research findings for comprehensive student performance prediction systems.

Purpose of the Study:

• To conduct a comprehensive literature review of student performance prediction.
• To analyze the association between student performance and influential factors from psychological and data analysis studies.
• To critically evaluate existing and novel prediction techniques and identify future research directions.

Main Methods:

• Systematic literature review of 1497 publications from 1990 to 2022.
• Analysis of psychological studies, data mining findings, and data analysis results.
• Comparative evaluation of student performance prediction techniques.

Main Results:

• Identified statistical associations between student performance and various influential factors.
• Evaluated the strengths and limitations of current and emerging prediction techniques.
• Highlighted opportunities for optimizing prediction systems through integrated approaches.

Conclusions:

• A comprehensive understanding of factors influencing student performance is crucial.
• Further research is needed to bridge the gap between psychological insights and data-driven prediction models.
• Future work should focus on developing robust, assumption-based datasets for improved prediction system optimization.