Academic Research Library

Find some of the best Journals and Proceedings.

XGBoost-Based Career Domain Prediction Using Academic Performance and Socioeconomic Features: An Empirical Study / Women in Buddhism: A Systematic Review of Their Entry, Role, and Empowerment from Ancient to Mauryan Period

Author : Kirti Khare, Dr. Anupa Sinha

Abstract : Accurate career prediction for students can significantly improve educational outcomes and reduce graduate underemployment. This paper presents an empirical study comparing five machine learning classifiers — Naive Bayes, Decision Tree, Logistic Regression, Random Forest, and XGBoost — for predicting student career domains across six categories: Technology, Healthcare, Business & Finance, Arts & Humanities, Law & Administration, and Engineering. A dataset of 750 student records was constructed using a structured survey collecting academic performance scores, technical and interpersonal skill ratings, and demographic attributes from undergraduate and postgraduate students in Chhattisgarh, India. Feature preprocessing included one-hot encoding, min-max normalization, and SMOTE-based class balancing within 10-fold stratified cross-validation. XGBoost achieved the highest accuracy of 81.3% with F1-score of 0.807 and AUC-ROC of 0.924. Feature importance analysis identified CGPA, Technical Skills Score, and Interpersonal Skills Index as the top three predictors. These findings confirm gradient boosting ensemble methods as the optimal approach for multi-class career prediction in heterogeneous Indian student populations.

Keywords : XGBoost, Career Prediction, Academic Performance, Machine Learning, SMOTE, Indian Higher Education, Feature Importance, Random Forest, Educational Data Mining.

Conference Name : International Conference on Data Analytics for Emerging Technologies and Applications

Conference Place : Pondicherry, India

Conference Date : 25th Jul 2026

Preview