ASSESSING A CANDIDATE'S SENIORITY LEVEL IN COMPUTER SCIENCE FIELD BY INTEGRATING SEMANTIC WEB TECHNOLOGIES WITH AUGMENTED REALITY

Usually, the process of recruiting employees is very complex, and this requires the allocation of numerous resources within the organization. Several steps of this process can be automated, especially related to Curriculum Vitae (CV) analysis. This paper presents the prototype of an application that combines semantic technologies with augmented reality (AR) in order to enhance the plain text from a CV content with context-aware information in regards to the technical background of the applicant. This approach proves to be more trustworthy than manual selection performed by the HR personnel, since rather than looking on an applicant profile as a collection of skills we try to reveal how all these competences interact with each other and what implicit knowledge the candidate has. We tested our prototype on a set of real CVs, gathered several metrics - number of found skills, number of knowledge graphs, number of significant knowledge graphs, maximum depth of known skills and primary known skills weight - and used them to analyze what influence they have on assessing the candidate's seniority level. We applied logistic regression with Bayesian approach and used the ROC curve in order to estimate the performance of the forecasting model.