Intelligent Resume–Job Matching Framework Using AI Techniques and Visual Analytics for Automated Recruitment

Abstract

The challenges that modern-day putting in place the recruitment process are huge due to the high number of applications to sieve through, the lack of efficiency in the screening process and manual screening, evaluation inconsistency, and subjective biases. Current automated systems tend to be shallow of key-word matching, lack any semantic flexibility, job-specific adaptation, and built-in visual analytics, thereby restricting their utility in assisting data-driven hiring. In order to fill such gaps, this paper proposes the design and development of an Intelligent Resume Profiling and Matching System (IRPMS), a web-based program, which automates the resume-job match process by incorporating Natural Language Processing (NLP), machine learning, and visual analytics. The system is built as a Flask module-driven framework, where NLP pipelines composed of tokenization, lemmatization, entity detection and semantic similarity processing are used to process PDF and DOCX resumes. Candidate suitability is measured using a weighted scoring system, as the required skills are compared against preferred skills against a curated database and recruiters are able to apply customization in the weights to ensure that the item reflects the priorities of the organization. The outcomes are presented in an interactive dashboard format in forms of radar charts, visualizations of skill gaps and ranked lists of candidates, which increases efficiency, fairness and illumination. The methodology had four steps that are feasibility validation, system design, modular implementation and evaluation. Section identification (92%), skill extraction (88%), average processing times of just under five seconds per resume, support of batch upload and strong error tolerance were also noted. Limitations consist of reliance on quality of job description and limits of the database. The aspiring applications will include the area-specific taxonomies, OCR, real-time labour market data and explicable AI. In general, the IRPMS is scalable, reliable, and bias-free solution that has dramatic potentials on contemporary recruitment activities.