Conference Paper

Behaviorally-Based Textual Similarity Engine for Matching Job-Seekers with Jobs

Heggo I.A.
Abdelbaki N.

Understanding both of job-seekers and employers behavior in addition to analyzing the text of job-seekers and job profiles are two important missions for the e-recruitment industry. They are important tasks for matching job-seekers with jobs to find the top relevant suggestions for each job-seeker. Recommender systems, information retrieval and text mining are originally targeted to assist users and provide them with useful information, which makes human-computer interaction plays a fundamental role in the users’ acceptance of the produced suggestions. We introduce our intelligent framework to help build the knowledge required to produce the most relevant jobs based on processing each job-seeker profile’s text, the behaviorally collected text and the jobs’ profile content. We analyzed the available textual similarity scoring algorithms to find the best suitable relevancy ranking model which is plugged into our developed textual similarity engine. The main purpose is enhancing the recommendation quality in the challenging domain of e-recruitment by finding the textually similar jobs for each job-seeker profile. © 2018, Springer International Publishing AG.