Topic Modeling Driven Content Based Jobs Recommendation Engine for Recruitment Industry
Shivam Bansal, Aman Srivastava, Anuja Arora · Procedia Computer Science · 2017
A number of postings for different job roles and job positions are posted at numerous sources in the recruitment industry. Therefore, this is a challenging and time-consuming task to collate the information and find out most relevant user-job connection mapping according to the skills and preferences of a user. This research work has been done to cover up this same problem and efforts have been made to provide a feasible and efficient solution for the same. We suggest a content-based recommendation engine, which automatically provides best suggestions to users by matching their interests and skills with the features of a job posting. In order to produce an intended recommendation, the proposed engine applies various text filters and feature similarity measurements. Similarity techniques use the bag of n-grams and topic models as the elements of feature vectors. The validations and testing of the model on real data obtained from a top job posting website show the applicability and efficiency of using topic models as features. The approach is generic and can be replicated to different industries.