Abstract—This paper presents the design and implementation of a semi-automatic semantic tool for boosting the recruitment process. It is able to analyze a set of CVs in order to identify a suitable candidate for a particular job. The process is based on semantic processing of the resumes and matching the candidates’ skills with the requirements for each particular job description. The purpose is to reduce the recruiter’s processing time by eliminating certain repetitive activities in the resume analysis procedure and to obtain a qualitative improvement by highlighting the competencies and qualities of candidates based on complex and customized semantic criteria. Index Terms—data mining, semi-automatic tool, recruitment process, semantic technologies I. INTRODUCTION People are the most valuable asset of an organization. One of the key roles of HR specialists is to find and hire the right people for the right job. But how do we find them? Their profile must match the organizational vision and mission, their set of values must be convergent with the organizational culture and their skills must match the job description and requirements. The evaluation process during recruitment is the key to finding the appropriate candidates in a timely manner. At present, the recruitment process is mainly assisted by human experts, which can be a time consuming task. Moreover, we assist at a shift in in recruitment methods, from the traditional one to a social recruiting, where decisions are based on social media analysis and on networking decision factors [1]. Companies currently create and use networks of leads, candidates, employees and alumni in order to leverage and increase the quality and performance of the recruitment process. An automation we can bring into the process is a smarter way of processing candidates’ resumes. If we build an improved and more intelligent parsing and skills matching algorithm, the entire recruiting process would benefit and we would be able to efficiently identify the suitable Alexandra Cernian is with the University Politehnica of Bucharest, Faculty of Automatic Control and Computers, Bucharest, 060042, Romania (phone: +40 744 632 031; e-mail: Alexandra.cernian@aii.pub.ro). Valentin Sgarciu is with the University Politehnica of Bucharest, Faculty of Automatic Control and Computers, Bucharest, 060042, Romania; (e-mail: vsgarciu@aii.pub.ro). candidates for a particular job. The final objective is to minimize the HR specialist input in the process and limit the time they spend in processing resumes by developing digital tools to automate the candidates identification phase. Such a tool would bring two main benefits: 1. Eliminate certain repetitive activities in the resume analysis procedure, currently performed by HR specialists mostly empirically and 2. Improve the process from a qualitative point of view by identifying and highlighting the competencies and skills of candidates based on customized criteria [2]. This paper presents a semi-automatic tool for boosting the recruitment process through a semantic approach for identifying candidates skills and matching them against organizational requirements and job descriptions. The rest of the paper is organized as follows: Section 2 presents and overview of the system, emphasizing the main objectives and benefits of this approach, Section 3 presents the design and implementation of the tool and Section 4 draws the conclusions of the hereby presented work. II. OVERVIEW OF THE SEMI-AUTOMATIC TOOL FOR IDENTIFYING CANDIDATES SKILLS This objective of this paper is to present the design and implementation of a semi-automatic tool able to assist HR specialists in the recruitment process by providing an automatic analysis of the candidates resumes and identifying their key skills, by also matching their competencies against the requirements in the job description. The system will be able to recommend the best suited candidates for specific positions. The process is based on a semantic processing of the candidates resumes, based on a dictionary approach. The flow of the application is presented in figure 1. Figure 1. The main components of the tool Parse document Match score Recommend candidates Boosting the Recruitment Process through Semi-Automatic Semantic Skills Identification Alexandra Cernian, Valentin Sgarciu Proceedings of the World Congress on Engineering 2017 Vol II WCE 2017, July 5-7, 2017, London, U.K. ISBN: 978-988-14048-3-1 ISSN: 2078-0958 (Print); ISSN: 2078-0966 (Online) WCE 2017