Date of Publication :15th March 2025

Abstract: The present study involves the use of machine learning techniques, specifically the multi-layer perceptron (MLP) model, to analyze student placement success. With the requirement for skill-based hiring, a deep understanding of how self-learning facilitates employability requires further research on this topic. The present paper assesses predictive models based on student placement data that incorporate the use of online courses, workshops, and certification as self-learning engagements. The study is based on a dataset that consists of 24 attributes: academic performance, self-learning activities, and placement outcomes. Comparison of the machine learning models of Decision Trees, Random Forest, XGBoost, and deep learning approaches, such as CNN, LSTM, and MLP. The MLP model has shown high accuracy, which is 96models. Feature selection techniques were used to improve the reliability of the predicti on. The results show that self-learning activities are significantly related to placement success, and thus there is a need for structured self- learning programs in academic curricula. This study adds value to campus placement strategies and provides actionable insights for educators and students. Future work can integrate hybrid models that incorporate attention mechanisms to improve predictive capabilities. These results reinforce the necessity of continuous self-learning for students who aim for successful employment outcomes in competitive job markets.

Reference :