Enhancing Resume Analysis: Leveraging Natural Language Processing and Machine Learning for Automated Resume Screening Using KSA Parameters

In the digital age, the recruitment landscape has changed dramatically. This study aims to optimize recruitment by automating resume screening through advanced Natural Language Processing (NLP) and Machine Learning (ML) techniques. The purpose of this research is to address the limitations of traditional Applicant Tracking Systems (ATS) by introducing a framework employing Bidirectional Encoder Representations from Transformers (BERT). A comprehensive methodology was implemented, including data preprocessing, manual dataset annotation, feature extraction via TF-IDF, and cosine similarity for ranking resumes against job descriptions. Empirical findings reveal that BERT outperforms other ML technologies including SVM, achieving a classification accuracy of 78.6%, validating its effectiveness in KSA extraction and alignment. Additionally, the system's outputs were validated by industry experts, who confirmed the accuracy and relevance of the extracted KSAs in real-world scenarios. The practical implications of this work include providing actionable insights for HR professionals, enhancing candidate-job matching, and streamlining recruitment workflows. The originality of this study lies in integrating advanced ML models for dynamic KSA classification, paving the way for more efficient recruitment practices. Future research may explore ensemble techniques and transfer learning to further refine the methodology and adapt it to diverse industries.