Towards Automating the Human Resource Recruiting Process

Abstract

Companies often receive numerous resumes for each job vacancy, and sometimes the resumes are not classified or even relevant to the job. Consequently, it is a time-consuming task for Human Resources (HR) to shortlist the candidates. In this work, following business process re-engineering and replacing Artificial Intelligence (AI)-driven approaches with organizational processes, we aim at technology disruption by proposing a holistic approach for resume recommendation in recruitment systems. This is done by harnessing the power of novel Machine Learning (ML) algorithms to address the candidate ranking problem. The proposed system starts with a preprocessing phase to extract a set of information from PDF files. Next, it applies ML techniques to compute the similarity between the submitted resumes and the target job description. Finally, it ranks the job-seekers and recommends the best candidates to the human resource. To the best of our knowledge, this is the first work that focuses on the Persian language enabling HR to identify the resumes that are closest to the provided job description.