| It is always a dream of AI researchers to enable the computer tounderstand human emotion. Human emotion analysis is a technology thatintegrates computer sensors and pattern recognition, which will have broadapplication in our daily life in the recent future. It will play an important rolein health care, security, entertainment and Human-Computer Interaction areas.This work proposes a solution of recognizing learner’s emotion status throughphysiological signals under E-Learning environment, which is of upperimportance for solving the critical issues: the lack of emotional interactionbetween student and teacher. Making great use of AI and advanced methodsto use the state-of-art computer technologies to support a human centric,intelligent education environment for teachers and students over the networkand physical classrooms, will sure to be the future direction of E-Learning, inwhich the emotion computing methods has been widely popularized. Thephysiological-based emotion detection method, a key part of emotioncomputing, enables the reflection of physiological signal aroused by differentemotions as a more robust and visible way of detecting emotion comparedwith other detection methods.The main contribution of this article is building up aphysiological-oriented emotion detection software system based on the theCRBM+SVM solution, proposed by the E-Learning Lab of SJTU. The paperstudied and compared physiological and psychological materials and researchmethod, combining a set of hardware sensor used for signal recording on thebase of existing psychological research findings. The paper describes the design and construction of a prototype system including the global functionmodule defining the main frame, the hardware interaction functionmodule,data transaction module,processing and calculate module, thatrealize the physiological-based emotion detection method to validate theeffectiveness of the existed solution and to build a base framework for futureapplications. |
PDF Full Text Request