Knowledge extraction from online vacancies for effective job matching
Sisay Adugna Chala, Scott Harrison, Madjid Fathi · 2017
Recruiters evaluate and filter job seekers, ranking them on various criteria. This includes how much of the required and desired requirements are satisfied, ensuring the candidate is the “best match” to vacancy. However, most vacancies do not classify the set of skills as required and desired explicitly. Required skills are those skills a job seeker must have in order to be considered for the job. Desired skills, on the other hand, are those skills which are useful but not essential for the job. The problem is that in automated matching systems avoiding trap of job seeker satisfying desired requirements to excel those who fulfill essential requirements. In this paper, we discuss the approaches of vacancy extraction and representation based on required and desired criteria, with the objective of matching them to job seekers with these criteria. We show how plain keyword-based vacancy-to-jobseeker matching (without taking into account these two types of criteria) may result in improper matching.