This research paper presents an AI-powered resume analysis system that improves traditional Applicant Tracking System (ATS) screening using Natural Language Processing (NLP) and semantic similarity techniques. The proposed system utilizes RoBERTa transformer models to evaluate contextual similarity between resumes and job descriptions instead of relying solely on keyword matching. The framework also integrates Named Entity Recognition (NER)-based skill gap analysis to identify missing competencies and provide actionable recommendations for candidates. The system supports secure PDF resume processing, ATS-compatible scoring, and interpretable insights through a user-friendly interface. Experimental observations demonstrate that semantic similarity-based evaluation provides more accurate and context-aware candidate assessment compared to conventional keyword filtering methods. Keywords: Resume Analysis, NLP, RoBERTa, Semantic Similarity, Applicant Tracking System, Skill Gap Analysis, Artificial Intelligence.
Shaikh et al. (Sun,) studied this question.