“AI-Driven Semantic Matching and Skill Gap Analysis: A Transformer-Based Approach using SBERT
AJITH NAIDU and PRATHAMESH AMBRE
Design and Implementation of an AI-Driven Skill Gap Analysis System, which utilizes Sentence-BERT (SBERT) and the all-MiniLM-L6-v2 architecture to automate and enhance the precision of resume-to-job matching. This study transitions from traditional keyword filtering to deep semantic similarity, allowing for the quantification of talent alignment and the detection of specific technical competency gaps across multiple job descriptions simultaneously. By integrating an Analysis Terminal that visualizes these data points in 3D, the project demonstrates a scalable “Secondary Research” approach to bridge the divide between academic qualifications and industry requirements, ultimately providing personalized upskilling roadmaps for career advancement.