A Rough Set and SVM Based Approach to Chinese Textual Affect Sensing

Text is an important modality for human-computer interaction, so studying the relationship between natural language and affective information as well as assessing the underpinned affective qualities of natural language has been the focus of research community. Several approaches have been performed to sense affect from English text, but the study on Chinese text emotion detection is still at the beginning. In this paper, we devote ourselves to sensing affective information from Chinese documents with the aim to group those into a set of emotions. A rough set and SVM based approach is adopted to categorize text into four emotional classes, including happy, sad, anger and surprise. Meanwhile, a Chinese textual emotion database is established to assist the processing.