Abstract— Job seekers, particularly fresh graduates and early-career professionals, often struggle to align their resumes with desired job roles due to limited awareness of industry expectations and personal skill gaps. Existing recruitment systems largely rely on keyword-based matching, which fails to capture deeper semantic relationships between candidate experiences and job requirements. This paper proposes an AI-driven resume matching and recommendation framework that addresses these limitations through semantic understanding and structured analysis of resumes and job descriptions. The proposed system adopts a multi-stage pipeline comprising structured information extraction, semantic representation learning, similarity computation, and personalized feedback generation. By transforming unstructured resume and job description text into structured representations, the system enables robust semantic comparison beyond surface-level keyword overlap. Advanced similarity scoring techniques are employed to evaluate alignment between candidates and job roles, resulting in an interpretable Career Fit Score. Additionally, the system identifies missing or weak skill areas and provides tailored recommendations to support career development and role readiness. The proposed framework contributes to the development of intelligent recruitment and career guidance systems by offering a scalable, explainable, and semantically informed solution that benefits both job seekers and recruiters in the evolving employment landscape. Keywords— Semantic Similarity, Job Recommendation Systems, Natural Language Processing, Skill Gap Analysis, Career Guidance Systems.
A et al. (2026) studied this question.