Azade Mohammadi - Department of computer engineering and IT, Razi University, Kermanshah, Iran
Mohammad kazemifard - Department of computer engineering and IT, Razi University, Kermanshah, Iran
Jahangir karami - Department of Psychology, Razi University, Kermanshah, Iran

In recent years, there has been a growth in the use of touch input devices. So emotion detection of interactive system users without any extra hardware are a very attractive field. On the other hand, E-learning has become a popular paradigm. One of the most important situations in E-Learning is multiple-choice test. The level and quality of the user’s tactile behavior in this situation is very different from others such as online games and we can’t use the research results in this situation, so a special research is needed for this situation. A good system may not only recognize users’ emotions, but also react accordingly to the detected emotions. In this research, we first propose an approach to detect three common emotions in learning process, we focuses on negative emotions: stress, frustration and boredom. we defines a features set and then NaiveBayes algorithm is used for classification. After that we use the detected emotions and personality type to react to the user. In this paper the Enneagram methodology, with the RHETI test is used to determine the user personality type. The results show that the combination of user emotion and personality type is very effective in improving user performance.

emotion detection, E-Learning, touch screen, personality type, Enneagram