Machine Learning for Student Placement Forecasting: An Empirical Study with ANN and Classical Classifiers

Accurately forecasting student placement outcomes has become increasingly essential for academic institutions striving to bridge the gap between educational programs and industry expectations. As institutions seek to promote career readiness and meet employment standards, the ability to comprehend the complex, multidimensional factors influencing student employability has proven to be a significant challenge, mainly when relying on conventional assessment methods. To address this issue, the current study compares the effectiveness of a customized Artificial Neural Network (ANN) model with several well-established machine learning classifiers. The suggested ANN is designed to capture complex interactions among several student-related characteristics, in contrast to existing approaches. The study employs a real-world dataset that contains a range of student behavioral and academic characteristics. Four traditional algorithms—Logistic Regression, Support Vector Machine (SVM), Random Forest, and K-Nearest Neighbors (KNN)—are examined in conjunction with the ANN model for assessment. A flawless F1-score, a recall rate of 99.89%, a perfect precision of 100%, and an accuracy of 99.65% show the ANN's distinct performance advantage.