categorizingemotionalvocabulariesinchinesenatural.doc

Categorizing Emotional Vocabularies in Chinese Natural Language Communication Xiaoyin Xu School of Chinese Language and Literature Beijing Normal University Beijing, China, 100875 xuxiaoying2000@ Jianhua Tao National Laboratory of Pattern Recognition Institute of Automation, Chinese Academy of Sciences, Beijing, China, 100190 jhtao@ Abstract Based on the brief review of some psychological theories of emotions and the introduction on our model for emotion parsing: Model of context based emotion detection from the text input, this paper constructs a Categorizing Emotional Vocabularies, which can be classified into emotional keywords, metaphor words and modifier words. The main categories and the sub-categories of the emotional vocabulary keywords will be introduced and the general syntactic rules of emotion vocabularies, for instance, influences in the syntactic structures or the sequences of part of speech (especially adjectives, adverbs and verbs) will be analyzed further. 1 Introduction With the development of computer technologies, more and more text message has been widely broadcasted in the internet. Computer science recently began, with success, to endow natural language dialogues with emotions. (Ortony, Clore Collins, 1988; Lazarus, 1991), such as the emotion description systema (Liu, Selker Lieberman, 2003), automatic chat system (Zhang, Barnden, Hendley Alan, 2006), emotional text-to-speech conversion (Tao, 2004; Campbell, 2001; Schroeder, 2006), and so on. To get the emotion state from text, some cognitive models have been tried (Carofiglio, de Rosis Novielli, 2008; Masum Shaikh, Prendinger Ishizuka, 2008). Among the technologies have been used, The categorizing of emotional vocabularies plays important roles in the research of textual emotion parsing. However, till now, the resources for emotional vocabulary are still very limited. Most of the previous work was just focused on the collection of limited emotional words. For instance, Whissell