Job Recommendation System, Machine Learning, Regression, Classification, Natural Language Processing

In today's highly competitive job market, it is becoming increasingly important for companies to hire employees who are best fit for a job and to ensure they retain these employees in the long run. Studies have shown that employees who find their job meaningful and satisfying are generally more productive and less likely to leave the job. Human Resource professionals therefore need to ensure that proper screening of candidates is conducted during the recruitment process and that they hire the best fit candidate for a job. Given the usually high number of applicants for a particular job, the recruitment process is time consuming and it is not always possible to conduct proper screening and interviews for each applicant. This paper presents the development of JobFit, a job recommendation system which makes use of a recommender system, machine learning techniques and past data to predict the best fit candidate for a job. The proposed job recommendation system takes as input the requirement of a job and the profile of the applicants and outputs a JobFit score indicating how fit each applicant is for the particular job. The system ultimately provides the HR professionals with a sorted list of all candidates with those who are more fit and apt for the job recommended first. This shall help to ensure the HR focus on the screening and interviews of only a small pool of candidates, the best ones recommended by the system, while at the same time be confident that the better candidates are not being missed.