Predictive Analysis in College Placement: Enhancing Fresh Graduates Employability with Machine Learning

Abstract

To improve the employability of recent graduates, this research article investigates the incorporation of machine learning and predictive analysis in college placement. Recent graduates need help in the modern employment market, and traditional placement methods must adequately match graduates' abilities with industry needs. Utilizing data-driven methodologies to forecast patterns and enhance the placement procedure, this study tackles this problem. The background emphasizes young graduates' challenges in finding suitable jobs, underscoring the necessity of creative placement tactics. The inefficiencies in the current placement processes, which result in a mismatch between employers' expectations and graduates' talents, are highlighted in the problem statement. The study's objective is to utilize machine learning and predictive analysis to align graduates' skill sets with the evolving demands of the labour market. The study comprises a literature review examining previous research on employability and college placement, a theoretical framework, methodology details, and data analysis results. Future study directions are suggested, difficulties and constraints are addressed, and results are interpreted in the discussion. This study intends to use predictive analysis to offer valuable insights for creating college placement methods to help graduates and companies looking to hire them. Here, we have proposed a comparative study about placement Prediction that will be performed using four different machine learning algorithms, including XGB Classifier, Decision Tree Classifier, K Neighbors Classifier, and Random Forest Classifier and acquired accuracy of 88.67%, 87.60%, 84.23%, 86.79% respectively. Using fine model tuning and hyperparameters may increase accuracy for the same datasets.