期刊名称：International Journal of Computer Science and Network Security
印刷版ISSN：1738-7906
出版年度：2010
卷号：10
期号：5
页码：208-214
出版社：International Journal of Computer Science and Network Security
摘要：Various emotions such as happy, sad, angry, surprised, and disgust have been known for a long time and become an important aspect of human behavior. However, the application of emotion has not been used in human and computer interaction. Therefore a good system in human interaction and computer should be able to recognize, interpret and process human emotions. Research in the field of emotion is a complex process because it can change dynamically. Because of that, the study of text-based emotion mostly done due to text forms is relatively simple compared to other forms such as visual or sound. Various methods has been used to make the process of recognition of human emotions, such as SVM (Support Vector Machine), VSM (Vector Space Model), and Cauchy Naive Bayes. As for the process to produce emotional responses, method that have been used is Fuzzy Logic. In this research will be discussed visualization of emotional expression on the face of three-dimensional model using Naive Bayes and Fuzzy logic. A combination of both methods can produce emotional expressions of an Indonesian-language text. Results of Naive Bayes text classification accuracy for yield 64.83%. While the results of a Fuzzy Logic facial parameter values are dynamic, so they can show facial expressions that contain more than one type of emotion. From the distribution of questionnaires to obtain accurate values for facial expression 66.4%.
关键词：Text classification; Naive Bayes; Fuzzy Logic; 3D Face Model
Loading...
