Resume Screening Using KNN, Random Forest Classifier and DistilBERT

Abstract

One of the most significant and critical tasks for every firm is to find the right person for the position. As online recruitment becomes more prominent, conventional hiring practices are becoming inefficient. Conventional approaches typically consume more time due to manually reviewing all applicants, assessing their resumes, and then creating a list of candidates who should've been interviewed. Many company hires other firms to screen their candidates resume and find out the suitable person for the position. In this information age, job searching has become both smarter and easier. Companies get a lot of resumes/CVs, and many of them aren't well-structured. Finding suitable candidate for any position takes a significant amount of time and effort. In this study, we have come up with an easy and effective solution for this tedious work. We build three models KNN, Random Forest Classifier and DistilBERT on same dataset for resume classification process. KNN and Random Forest Classifier model have achieved highest accuracy 98% among all the models.