Predicting Student Placements Using Machine Learning Algorithms

Abstract

Predicting student placements is a critical task that can significantly impact the career trajectories of graduates and the efficiency of educational institutions' placement processes. In this research work, we investigate and study the use of machine learning techniques. to predict student placements using a comprehensive dataset containing attributes such as well as factors including age, gender, stream, internships, CGPA, lodging, and prior shortages. The study commences with data collection and preprocessing, ensuring data integrity and suitability for machine learning tasks. The process of enhancing features is used to extract meaningful insights from the data, and exploratory data analysis provides valuable visualizations and descriptive statistics, revealing underlying patterns and trends. Four machine learning algorithms, namely Decision Tree, Random Forest, XGBoost, and K Nearest Neighbors (KNN) classifiers, are employed for model training. The models are evaluated using various performance indicators to evaluate their ability to forecast, such as accuracy, precision, recall, and F1-score capabilities. Results indicate that the Random Forest Classifier stands out as the most effective model, achieving the highest accuracy and F1-score among the evaluated models. It demonstrates robust predictive capabilities and accurately classifies students into placed and not-placed categories. The study underscores the significance of predicting student placements, benefiting both educational institutions and students in making informed decisions. By understanding the factors influencing placement outcomes, institutions can optimize their placement programs, providing better career opportunities for students.