Machine learning analysis of factors affecting college students' academic performance

Area of Science:

• Educational Psychology
• Data Science in Education

Background:

• Academic performance is influenced by multifaceted factors.
• Understanding these factors is crucial for effective educational strategies.

Purpose of the Study:

• To identify key factors impacting college students' academic performance.
• To evaluate the predictive power of machine learning models for academic success.

Main Methods:

• Chi-square tests were used for feature selection.
• Machine learning models including Logistic Regression (LOG), Support Vector Classification (SVC), Random Forest Classifier (RFC), and XGBoost were employed for prediction.

Main Results:

• The XGBoost model demonstrated superior performance in recall and accuracy.
• Metacognitive awareness, learning motivation, and participation in learning were identified as critical factors.
• Time management, environmental factors, and mental health also significantly impact academic achievements.

Conclusions:

• Metacognitive awareness, motivation, and participation are vital for academic success.
• Professional training positively influences academic performance by integrating theory and practice.
• Educational guidance should focus on enhancing these factors and supporting student well-being.