Design and Implementation of a Smart NLP-Based Resume Screening Tool for Technology-Driven Recruitment

Abstract

Modern recruitment increasingly depends on digital ecosystems in which speed, scalability, and objectivity are primary drivers of competitive advantage, and organisations routinely review large numbers of resumes for a single vacancy. Traditional resume screening is time-consuming, inconsistent, and highly dependent on manual judgment or simple keyword matching, which may cause qualified candidates to be overlooked when their resumes use different wording from the job description. This study presents the design and implementation of a smart Natural Language Processing (NLP)-based resume screening tool that supports automated resume analysis and candidate ranking. The system extracts resume text, preprocesses the content, identifies key information such as skills, education, and experience, and compares resumes with job descriptions using keyword matching and semantic similarity techniques. A weighted scoring mechanism combines both signals to generate a final relevance score, allowing candidates to be ranked according to their suitability for a given job role. The system was evaluated based on matching accuracy, ranking consistency, processing efficiency, and result interpretability. The findings indicate that the proposed system improves screening consistency and reduces manual effort while maintaining transparency through explainable scoring, directly addressing concerns about algorithmic bias and the lack of explainability reported in technology-driven recruitment. Therefore, the system provides a practical and scalable solution for supporting early-stage recruitment processes, particularly for small and medium-sized organisations seeking context-aware yet interpretable screening.