
Research Article
CVO: Curriculum Vitae Optimization by Recommending Keywords to Undergraduate Students
@INPROCEEDINGS{10.1007/978-3-031-33614-0_19, author={Cibele Santos and Fabr\^{\i}cio G\^{o}es and Carlos Martins and Felipe da Cunha}, title={CVO: Curriculum Vitae Optimization by Recommending Keywords to Undergraduate Students}, proceedings={Big Data Technologies and Applications. 11th and 12th EAI International Conference, BDTA 2021 and BDTA 2022, Virtual Event, December 2021 and 2022, Proceedings}, proceedings_a={BDTA}, year={2023}, month={5}, keywords={Clustering Curriculum Vitae Optimization Keywords Recommendation}, doi={10.1007/978-3-031-33614-0_19} }
- Cibele Santos
Fabrício Góes
Carlos Martins
Felipe da Cunha
Year: 2023
CVO: Curriculum Vitae Optimization by Recommending Keywords to Undergraduate Students
BDTA
Springer
DOI: 10.1007/978-3-031-33614-0_19
Abstract
Candidate selection platforms have been widely used in companies that seek agility in the process of hiring. Candidates who do not meet the requirements of a job vacancy are disqualified in the first step, called screening. This stage has been automated due to the large volume of curriculum vitae (CV) of candidates per vacancy, particularly for internship vacancies. As a consequence, candidates receive little to none feedback and do not know how to improve/optimize their CVs for new applications. The goal of this paper is to realize the curriculum vitae optimization (CVO) process for internship vacancies by implementing a recommendation system that given an undergraduate student CV, it suggests the addition of relevant keywords, taking into account the student’s undergraduate course. This system is implemented based on the clustering of CVs keywords, from an internship recruitment private company database, into profile groups which are linked to internship vacancies. The experimental results showed that recommendations improved students CVs similarity (competitiveness within a specific field) from 18.83%, with 3 keywords recommendation, up to 50.67%, with 10 words.