Analyzing Key Determinants of On-Campus Placement Using Machine Learning

 Keerthiraj U R

PG Department of Computer Applications JSS College of Arts, Commerce and Science, Mysuru, India

Mrs. Shwetha M R

Assistant Professor, PG Department of Computer Applications

JSS College of Arts, Commerce and Science, Mysuru, India

Abstract— On-campus recruitment plays a decisive role in shaping the career prospects of graduating students, yet institutions frequently rely on subjective, score-only assessments that fail to capture the many attributes that actually drive hiring decisions. This paper presents a machine learning based framework for analyzing the key determinants of on-campus placement and for predicting individual placement outcomes. Student data spanning academic performance, technical proficiency, communication skills, internship exposure, certifications, and project experience is collected, cleaned, and transformed through a structured preprocessing pipeline before being used to train and compare three classification algorithms: Logistic Regression, Decision Tree, and Support Vector Machine. Beyond outcome prediction, the system performs feature importance analysis to rank the attributes that most strongly influence placement success, giving students and placement officers actionable, data-driven insight rather than a single opaque score. A web-based dashboard exposes the predictions, probability scores, and ranked determinants to administrators, faculty, and students. Experimental evaluation shows that ensemble and tree-based approaches achieve the highest predictive accuracy while Logistic Regression remains the most interpretable baseline. The proposed approach demonstrates that data-driven analytics can meaningfully complement traditional placement preparation by enabling early, targeted intervention for at-risk students.

Index Terms— campus recruitment, decision tree, feature importance, logistic regression, machine learning, placement prediction, support vector machine.