DROP OUT PREDICTION USING MACHINE LEARNING MODELS

ABSTRACT

This research project aims to design and implement a machine learning approach to assess the identity and motivation of undergraduate engineering students at the University of Benin's Faculty of Engineering. The study employs an exploratory mixed-methods approach, combining quantitative analysis through machine learning models and qualitative insights from existing instruments. Linear regression and deep learning models are utilized to analyze factors affecting student motivation, while a structured instrument captures relevant data on identities and motivations. The project involves a comprehensive PHP-based web application for data processing, sentiment analysis, and visualization. The findings are expected to contribute to targeted interventions, support strategies, and curriculum design to enhance student engagement, retention, and success in engineering programs. Challenges addressed include database integration, performance optimization, and ethical considerations. The research contributes to the field of computational psychology by providing insights into the complex relationship between identity and motivation, potentially improving enrollment and success rates in engineering education.