Affective Computing Based on Ontology and Fuzzy Inference

• Main Authors: Ren-ying Fang, 方仁穎
• Other Authors: Hao-Chiang Lin
• Format: Others
• Language: zh-TW
• Published: 2010
• Online Access: http://ndltd.ncl.edu.tw/handle/01831867297313490253

碩士 === 國立臺南大學 === 數位學習科技學系碩士班 === 98 === This study integrates the ontology and the fuzzy inference algorithm to create a new type of affective computing system. The system we supposed allows users to use common colloquial sentences to gain current mood quickly, it made people feel fun and interesting. Furthermore, we can release their stress when they are in bad moon. Since the quality of our life is getting down although the technology is getting blooming. We construct an Affective Computer which capable of emotion analysis and recognition. Base on the Affective Computing to find a way to relieved stress not only psychology but physical. The following are the system functions which we have developed: (1) The parser based on natural language processing. (2) Interactive semantic detection and recognition system. (3) OMCSNet Concept semantic knowledge databases for Affective Computing. (4) Fuzzy Inference and Entropy for classification of emotion. Intergrated of four features we have implemented allow the system to process emotion recognition for any sentence, and achieve 76% precision of the emotion recognition rate. By combining the concept of semantic knowledge, AI and neural network, we have a deeper exploration about Affective Computing.