Automating Recruitment Process Using NLP and Deep Learning: A Novel Approach for Accurate Candidate Selection

Abstract

In the current recruitment industry, selecting the most suitable candidate from a large pool of resumes within a limited time frame presents a significant challenge. Traditional information retrieval techniques often fall short in meeting the needs of both job seekers and employers. This paper aims to optimize the recruitment process in the Moroccan job market by leveraging natural language processing (NLP) and deep learning to automate and enhance the matching process. The proposed framework, evaluated on a dataset of 325 job descriptions and tested on 30 LinkedIn job descriptions, demonstrated that the combined Named Entity Recognition (NER) and Part-of-Speech (POS) model achieved a precision of 0.96 and a recall of 0.88, outperforming the standalone NER model. The use of implicit skills detection further improved the relevance of CV recommendations, resulting in more accurate and efficient candidate selection.