Csaba Kiss

Csaba Kiss is affiliated with the Faculty of Psychology and Educational Sciences at Hyperion University of Bucharest, Romania. He holds credentials as a polygraph expert authorized by the Ministry of Justice in Romania, and supervised polygraph examinations conducted with the Romanian Police involving convicted offenders for research purposes. His work bridges academic research in psychology with practical applications in forensic credibility assessment and deception detection.

Kiss has made significant contributions to the emerging field of machine learning applications in polygraph science. His 2023 scoping review analyzed 57 relevant papers examining neural network applications in scoring polygraph tests, providing a comprehensive overview of the existing research on this subject. The review found that neural networks have shown promise in improving the accuracy of polygraph tests, with some studies reporting significant improvements over traditional methods. In subsequent research, his team trained a Multilayer Perceptron Neural Network to predict deception scores in 400 offender datasets based on psychophysiological indicators, achieving an 80% correct classification rate. His research has examined comparative machine learning approaches, including Random Forest, Support Vector Machine, and Neural Network regression models for automated deception detection in polygraph examinations.

Kiss collaborates extensively with researchers Dana Rad from Aurel Vlaicu University of Arad and Nicolae Paraschiv from Petroleum-Gas University of Ploiești, forming a productive research team focused on applying computational intelligence methods to psychological assessment. Their collaborative work has been published in MDPI's Information journal, the International Journal of Computers, Communications & Control, and Acta Polytechnica Hungarica. He serves on the Scientific Board of the Hyperion Psychology Conference, contributing to the academic community in Romanian psychology.

Beyond polygraph research, Kiss's broader scholarly interests span decision-making under uncertainty, integrating perspectives from psychology and systems engineering, as well as artificial neural networks in decision adoption systems for psychological data interpretation. His interdisciplinary approach combines forensic psychology, psychophysiology, and computational methods to advance credibility assessment techniques and psychological measurement more generally.