Font Size: a A A

Emotional Models And Modeling Techniques In E-Learning

Posted on:2013-05-31Degree:MasterType:Thesis
Country:ChinaCandidate:J B WuFull Text:PDF
GTID:2358330371475554Subject:Computer application technology
Abstract/Summary:PDF Full Text Request
Research of this thesis supported by the National Natural Science Foundation of China(Emotional cognitive analysis and research based on studies expression in E-Learning),Beijing Natural Science Foundation(The Study of Personalized e-learning Community Education based on Emotional Psychology),Beijing Education Committee Science and Technology project and Beijing Natural Science Foundation(Key Technology of Personalized Digital Education based on cognitive emotion model).The main purpose of this thesis is solving the emotional absence of a common problem in the traditional teaching network, and adding academic sentiment analysis in the E-Learning. Specifically, for learners'expression feature information extracted by the project group of students, it is to distinguish the emotional reaction of the learners' current learning content and to draw the learner emotion state, and then to provide emotional function for the personalized E-Learning system.The main research results are as follows:Firstly, based on emotional and educational psychology theory in E-Learning, it is modeling study emotion recognition for full discussion and analysis on the learners' characteristics. All of these factors are combined with OCC emotion model, from the aversion degree, cheer degree and pleasure degree three dimensions.Secondly, in order to improve the efficiency and identification accuracy of emotion recognition, through analysis and compare of experimental samples, it creates a personalized mood model according to the relevant conclusions in a model of academic emotions and the research of graduate students. And it introduces of the interpupillary distance as a major factor of the aversion degree. So it is effectively improved the accuracy of emotion recognition.Thirdly, using the support vector machine (SVM:Support Vector Machine) network of small sample nonlinear fast learning and classification features, it has created the aversion degree, cheer degree and pleasure degree three-dimensional emotion model. The model provides the necessary basis of academic emotions for interactive teaching reasoning in E-Learning.Finally, combination of personalized and emotion model, using the JAVA programming language on the Eclipse platform, the prototype system has been got a good experimental effect.
Keywords/Search Tags:E-Learning, Emotion recognition model, Variable distribution ofestimates, Support Vector Machine, OCC emotion model
PDF Full Text Request
Related items