Adaptive methods for job recommendation based on user clustering
In today's world, recommendation systems are used to solve the problem of information overload in many areas allowing users to focus on important information based on their interests. One of the areas where such systems can play a major role is in helping students achieve their career goals by generating personalized job and skill recommendations. At present, there are many job posting websites providing a huge amount of information and students need to spend hours to find jobs that match their interests. At the same time, existing job recommendation systems only consider the user's field of interest, but do not take into consideration the user's profile and skills, which can generate more relevant career recommendations for users. In this work, we propose CaPaR, a Career Path Recommendation framework, which addresses such shortcomings. Using text mining and collaborative filtering techniques the system first scans the user's profile and resume, identifies the key skills of the candidate and generates personalized job recommendations. Moreover, the system recommends additional skills to students required for related job openings, as well as learning resources for each skill. In this way, the system not only allows its users to explore large amounts of information, but also expand their portfolio and resume to be able to advance their careers further. We experiment and evaluate the various recommendation algorithms with real-world data collected from the San Jose State University career center web site.