Preliminary results in evaluating the pleasantness of an interviewing candidate based on psychophysiological signals

Job openings are getting scarce, as the number of applicants are increasing. Evaluation of job interviews is a serious burden for the human resource departments. Automatic evaluation of candidates applying for jobs is an important domain open for improvement. Evaluation of applicant videos from social signals is possible but may not be enough to determine candidates’ suitability for jobs that require social stress. The interview itself is a social stressor, as widely known from the Trier social stress test. In this study we aimed to aid the candidate evaluation process, based on physiological signals collected from the facial area. More specifically, we tried to determine the pleasantness of video strips of an applicant based on the EMG, SCR, pupil and HR signals. A simple binary kNN classifier was able to determine the pleasantness with an accuracy of 73%