Machine Learning Approach for Automated Resume Screening and Candidate Shortlisting

The present study successfully developed an AI-based recruitment prediction system for candidate shortlisting using machine learning techniques, feature selection methods, and MATLAB GUI implementation. The study utilized an AI Resume Screening Dataset containing multiple candidate-related parameters, including years of experience, education level, skills match score, project count, resume length, and GitHub activity, to analyze and predict recruitment outcomes. Initially, Linear Regression and Random Tree models were implemented using all available input parameters to evaluate their predictive capability. The obtained results demonstrated that both models were capable of predicting candidate shortlisting outcomes with moderate accuracy. However, the models also exhibited certain limitations related to prediction spread and classification performance. To improve model efficiency and prediction capability, feature selection techniques such as F-Test and Minimum Redundancy Maximum Relevance (mRMR) were applied. The feature selection analysis identified Years of Experience, Education Level, and Skills Match Score as the most influential parameters affecting recruitment decisions. Using these refined input features, an improved Linear Model was developed, which achieved a high prediction accuracy of 94%. The results confirmed that feature selection significantly improves model performance by eliminating redundant variables, reducing dimensionality, and enhancing prediction reliability. The refined model demonstrated better classification capability and stronger agreement between predicted and actual recruitment outcomes compared to models developed using all input parameters.   In addition to predictive model development, a MATLAB-based Graphical User Interface (GUI) was successfully designed and implemented to provide an interactive and user-friendly recruitment prediction platform. The GUI enables users to input candidate details and instantly obtain prediction results along with probability scores, thereby improving accessibility and practical usability of the developed AI system. The proposed study demonstrates the effectiveness of integrating machine learning algorithms, feature selection techniques, and GUI-based deployment for intelligent recruitment decision support systems. The developed framework has significant potential for practical implementation in automated hiring systems and can assist organizations in improving recruitment efficiency, reducing manual workload, and supporting data-driven hiring decisions. Future work may focus on implementing advanced deep learning models, Natural Language Processing (NLP)-based resume analysis, larger real-world recruitment datasets, and cloud-based deployment to further improve prediction accuracy, fairness, scalability, and real-time recruitment automation capabilities.

·        Gan, C., & Mori, T. (2022). Developing an English resume corpus to improve AI-driven resume screening performance. Journal of Applied Language Technologies in Recruitment, 7(2), 59–79. ·        Gan, C., & Mori, T. (2022). Developing an English resume corpus to improve AI-driven resume screening performance. Journal of Applied Language Technologies in Recruitment, 7(2), 59–79. ·        Harris, L., & Thompson, G. (2024). Benefits and limitations of AI-powered resume screening: Case research insights. Journal of Intelligent Recruitment Systems, 14(1), 93–112. ·        Hosain, M. S. (2025). The use of Artificial Intelligence (AI) in the hiring process. Journal of Innovation and Knowledge, 10(2), 100612. ·        Lo, F. P.-W., Qiu, J., Wang, Z., Yu, H., Chen, Y., Zhang, G., & Lo, B. (2025). AI hiring with LLMs: A context-aware and explainable multi-agent framework for resume screening. arXiv Preprint arXiv:2504.02870. ·        Mohammad Khan, F., Gupta, R., & Sekhri, S. (2021). Automated bacteria colony counting on agar plates using machine learning. Journal of Environmental Engineering, 147(12), 04021066. ·        Nilizadeh, S. (2021). Vulnerabilities in text-embedding-based recruitment algorithms: An adversarial analysis. Journal of Algorithmic Fairness in Recruitment, 4(3), 101–118. ·        Patel, S., & Kumar, V. (2021). Reducing hiring bias through AI-enabled anonymized recruitment systems. International Journal of Ethical Hiring and Workforce Diversity, 9(1), 72–89. ·        Pisanelli, E. (2022). Automated resume screening as a strategy to reduce gender bias in hiring. Journal of Economic Behavior & Organization, 204, 357–372.