Machine Learning-Based Student Placement Prediction Analysis: A Data-Driven Approach To Enhance Employability

Student placement prediction is a crucial task for educational institutions and recruiters to optimize campus hiring processes. This study employs a data-driven machine learning approach to predict student placement outcomes based on academic performance, extracurricular activities, certifications, aptitude scores, and soft skills ratings. A dataset comprising 10,000 student records was analyzed, and multiple machine learning models were trained, including LR, SVM, DT, RF, KNN, and GNB. To address data imbalance, the SMOTE was used to ensure robust model performance. Feature selection identified CGPA, aptitude test scores, and placement training as the most influential factors. Among all models, RF achieved the highest accuracy of 79.72%, outperforming traditional statistical methods. Model performance was evaluated metrics and execution time analysis. The findings provide valuable insights for students, helping them understand the key factors influencing placement success, and assist institutions in refining training programs. This study demonstrates the effectiveness of machine learning in enhancing placement prediction accuracy and suggests potential future improvements using deep learning techniques for more precise outcomes.