Analyzing Socio-Academic Factors and Predictive Modeling of Student Performance Using Machine Learning Techniques

Understanding the factors that influence student performance is crucial for improving educational outcomes. Thus, this study aims to examine the impact of socio-economic and psychological factors on student performance, less is known about how students' personal attitudes and behaviors across different departments and activities correlate with their academic success. This study employs exploratory data analysis (EDA) to identify trends and relationships within the dataset. Machine learning techniques, such as K-means clustering and Long Short-Term Memory (LSTM) networks, are utilized to model and predict student performance based on their reported behaviors and preferences. The dataset is reduced using Principal Component Analysis (PCA) to enhance the clustering process. The findings suggest significant variations in academic performance based on departmental affiliation, gender, and engagement in certification courses. The LSTM model achieved an accuracy of 91% on the test set, demonstrating substantial predictive capability. However, the classification report reveals that while the model was highly effective in identifying the majority class (label 1), achieving a precision of 91% and a recall of 100%, it failed to correctly predict any instances of the minority class (label 0). The insights from this study could help educators tailor interventions to address the specific needs of students based on their behaviors and departmental affiliations, leading to more personalized education strategies and potentially improving academic outcomes.