Akademik Veri Yönetim Sistemi
Tezin Türü: Yüksek Lisans
Tezin Yürütüldüğü Kurum: Orta Doğu Teknik Üniversitesi, Türkiye
Tezin Onay Tarihi: 2018
Tezin Dili: İngilizce
Öğrenci: Ceren Uyanık Civek
Danışman: DİDEM GÖKÇAY
Özet:Human-computer interaction can be enhanced if emotional arousal of the user can be predicted. Measurement of pupil dilation is an effective indicator to achieve a successive classification for categorizing the psychological state of a user. In this study, rather than trying to identify several psychological states, we focused on the identification of stress. There exist several factors that shift the state of a computer user from relaxation to stress. In this study, we mainly focused on the cognitive factors that cause stress by increasing the difficulty level of a task. We also evaluated the effect of color on emotional responses. We hypothesized that assigning more difficult tasks and looking at a colored image produce higher pupil dilation signals than the signals in a neutral state. In order to evaluate the effectiveness of these factors, we conducted experiments including two phases by using TOBII T120 eye-tracker system. In the first phase, a baseline was constructed by showing neutral IAPS images to record measurements during neutral emotion. In the second phase, some modifications on stimuli were made to increase cognitive load. Pupil measurements collected during these experiments were used to train supervised classifiers for categorizing stressful versus neutral states of the computer users. Both collective and individual subject-based analyses were performed. Better classification results are obtained for individual subject-based classification.