Recommendation using job profile analysis
Driss Mhamdi, Mohamed Azouazi, Reda Moulouki, Mohammed Yassine El Ghoumari
Academic Editor: Youssef EL FOUTAYENI
Received |
Accepted |
Published |
31 January 2020 |
15 February 2020 |
10 March 2020 |
Abstract: Job seekers are suffering with the big amount of information to retrieve in order to find the suitable job. Recommendation systems are an efficient solution to address this issue. A recommender system is a sort of software with the capabilities to make item recommendations based on user preferences and profile attributes[1]. Different methods and algorithms are used to recommend relevant job offers to candidates such as the similarity functions based on the fuzzy logic’s operators [2] and the classification algorithms from Naïve Bayes to Neural Networks [3]. Furthermore, Topic modeling technique could be applied to represent job offers by a limited number of topics [4]. To recommend relevant jobs, we have used Natural Language Processing technique [5] consisting on parsing job offers and resumes in order to extract meaningful features such as job titles, technical skills and qualifications. Then, we build resulted data in the form of tables to facilitate the matching of features from both job postings and candidate resumes.