Optimizing Recruitment and Job Sourcing with a Multilayer Perceptron Classifier-Based Recommendation System for More Effective Hiring Decisions

الملخص

With the increase in technological development, most jobs have specific advantages and specifications, and the size of the offer and the number of those wishing to hire have increased significantly, and the pace of increasing the number of job managers has also increased in the direction of obtaining a distinguished employee who is ready to fill the desired job. This paper focuses on building a recruitment system that aims to serve both parties in the job equation, to serve the job seeker, and to serve the job managers in finding the right employee. The research of this paper lies in building a job recommendation system based on the skills written in the employee's CV, and then the system provides a suggestion of vacant jobs that the employee can go to. This paper also provides a short presentation on the traditional prevailing framework within an important institution, which is the Misurata Free Zone. Where initially, the system works to clean the information by expelling incomplete information, or those with missing or duplicate parts. Then, job recommendations are provided to targeted applicants based on their preferences or on what is determined by the Human Resources Management Office in the Misurata Free Zone. It uses different machine learning procedures, the results of which show that artificial intelligence systems give the highest accuracy in predictions, especially when compared to traditional systems. The recommendation framework based on multiple attributes including geography is used to find suitable employees who are geographically close to the organization, which can help job seekers reach their destination at no cost. This system also reveals how much important data is lost in the recruitment process that is done incorrectly by traditional methods. It lays a clear and effective scientific foundation for reaching a suitable recommendation for both hiring parties.