Streamlit-Based Application for Predicting Job Prospects After Graduation using Logistic Regression

Abstract

This study presents the development of a web-based application designed to predict the likelihood of university students securing employment after graduation using logistic regression. The application was built with the Streamlit framework and processes input data such as GPA, internships, projects, certifications, soft skills, aptitude test scores, and participation in training or extracurricular activities. The model was trained using a dataset sourced from Kaggle, which reflects various academic and experiential attributes of students. After preprocessing and model training, the logistic regression model achieved an accuracy of 85%, with additional evaluation metrics indicating strong predictive performance. The application features real-time prediction, visual categorization of employment probability, and factor contribution analysis. The results show that aptitude test scores, GPA, and soft skills are the most influential factors in determining employment outcomes. This tool serves both predictive and educational purposes by providing career-related insights for students and decision-making support for academic institutions in planning targeted development programs.