Font Size: a A A

Research On Emotional Computing In The E-learning System

Posted on:2013-06-25Degree:MasterType:Thesis
Country:ChinaCandidate:W J WuFull Text:PDF
GTID:2358330371975673Subject:Computer applications
Abstract/Summary:PDF Full Text Request
In recent years, the application of E-learning system has made considerable progress and is gaining more and more attention, playing a great role in teaching. There are emotional problems which Traditional E-learning system has, such as lacking of emotional interaction between teachers and students. By Introducing Affective computing module into the traditional E-learning system can effectively solve emotional loss problem, and personalize teaching.This paper introduced the affective computing module into the E-learning system based on the research of the traditionalE-learning system. The module used facial expression as the main source of emotionto build the POMDP-based emotional model.The emotion model based on the learners'different emotional states makes emotional adjustment strategy to achieve the emotional interaction between the learner and the system.The core technology of the affective computing is emotion modeling based on POMDP and expression recognition technology.The worksmainly done in this paper are as follows:(1) Face Detection. This paper proposed a real-time method to detect faces in videos. This method firstly uses frame difference method to extract the motion area. The clustering character of skin is applied to get the general face area. Then we employ Adaboost algorithm to make concrete detection of human face. At last an improved CamShift algorithm method is used to keep the tracking and improve algorithm speed. The experiments demonstrated the robustness and high speed of the proposed algorithm.(2) Facial Expression Recognition. This paper presented an improved facial expression recognition method based on2DPCA and integrated neural networks. At first use2DPCA method to extract facial expression characteristics, then train classifiers using integrated neural network whose individual networks generated the integrated output were chosenby genetic algorithms. The experiments proved the effectiveness of the proposed method.(3)The Affective Computing Based on POMDP. This Paper proposed the affective computing method ofE-learning system based on the POMDP model.By understanding the emotions of learners,the system can make the real-time adjustment of teaching strategies.
Keywords/Search Tags:E-learning, Affective Computing, Face Detection, Expression Recognition, POMDP, Emotional Regulation
PDF Full Text Request
Related items