A Novel Job Recommendation System using Natural Language Processing

Abstract

The number of people graduating is increasing every year. The number of applicants for each job posting is also increasing. The task of recruiters is to go through the resumes, find suitable candidates, shortlist them, and schedule interviews, which could be time-consuming task. This task must be automated as much as possible so recruiters can focus on other tasks. This study proposed a Deep Learning approach to extract information from unstructured resumes and job descriptions using Natural Language Processing (NLP). Similar to a recruiter, it can automatically extract information from unstructured data and then use the extracted information to link each candidate to a job posting and then rank the candidates based on the job description. The process was accelerated by using the Streamlit package to rapidly deploy the model and create visualizations to make each job’s candidate recommendation process easier and more user-friendly. Our proposed approach allowed us to compare the shortlisted candidates in depth and rank candidates accordingly