A hybrid two-way recommender system for candidate and job search

The recruitment industry faces information overload when matching qualified individuals with competitive jobs. A recommender system can effectively solve this problem by filtering relevant content in order to make recommendations, in this case matching candidates with jobs. The objective of this project is to develop such a system for two-way search and ranking of candidates given job description and vice versa; put simply this system will recommend candidates to employer and will recommend job opportunities to candidates. Existing recommender systems have their own unique benefits. This work will build upon earlier research in this field by combining the enhanced interpretability and superior performance of Knowledge Graph recommender systems with the inferencing ability of Fuzzy Logic recommender systems, ultimately developing a two-way, hybrid recommender system with increased accuracy and explainable recommendations—an innovation that will provide competitive advantage in the recruitment industry.

Faculty Supervisor:

Uzair Ahmad

Student:

Ashish Kutchi;Pruthvi Ignole;Sukhpreet Kaur

Partner:

Eggdemy Inc.

Discipline:

Business

Sector:

Professional, scientific and technical services

University:

Durham College