Academic journal article Educational Technology & Society

Design and Implementation of Technology Enabled Affective Learning Using Fusion of Bio-Physical and Facial Expression

Academic journal article Educational Technology & Society

Design and Implementation of Technology Enabled Affective Learning Using Fusion of Bio-Physical and Facial Expression

Article excerpt

Introduction

In the age of Internet, the tutoring systems have transformed very rapidly. We have given shape of an organization which is proposed to play a complementary role in education by maintaining quality standard at all possible levels using all possible formats of e-education stressing on e-tutoring. Remembering their school and university years, very many people will agree that their emotional state at a lesson or at a test influenced the result of the process. On the scientific level, a fundamental contribution to clarifying the role of positive emotions of the students in the success of learning/teaching process was done by the System of Emotional-Imaginative Teaching, or the EIT-system (Fomichov, & Fomichova, 1997; 2006; Fomichova, & Fomichov, 1996; 2000). For the best tutoring, learner's current states need to be identified where the key solution is learner's emotion recognition. Without addressing of emotion feedback, tutoring system is imperfect. Emotions can have enormous effects on learning and play a vital role in decision making; Inclusion of emotion is must in e-education rather e-tutoring system.

The effectiveness of E-learning (Shen, Wang, & Shen, 2009) also depends on establishing two-way communication between mentors and learners, and among learners themselves. Our research objective is to incorporate new paradigm, that is, learner's emotion capturing and inclusion in conventional E-Learning System, such emotion based E-Learning system is called Affective E-Learning (Sandanayake, Madurapperuma, & Dias, 2011), to get more effective results. In this context learning is viewed as an active, subjective and constructive activity where the learner is at the center of the learning process (Kordaki, 2004). The major changes in our understanding of the nature of learning imply Technology Enabled Blended Learning (Webb, 2014). The objective of our work is to design a complete delivery model where instant learner's feedback can be captured at the client site through bio-physical and facial expression, which gets communicated to the e-mentor end. This helps to achieve efficient automatic delivery of lesson according to the learner's need enhancing the productivity and efficiency of E-Learning. For capturing bio-physical signals we have used three attributes Heart Rate (HR), Skin Conductance (SC) and Blood Volume Pressure (BVP) directly from the learner using ProComp5 device and is passed through the DTREG simulator with emotion-detection function and followed by the generation of the confusion matrix. Concurrently, we adopted the direct capturing of facial expressions using spot detection technique using Webcam from learner and generated the confusion matrix. Finally the two confusion matrices are fused at the decision level and we obtained the resultant confusion matrix. The result is passed through an automatic lesson detection algorithm to detect the lesson as per the learner's need. Designing of such automated system is a very challenging task, as all the detections have to be online at the learner end (client site).

Affective e-learning model

In the section we discuss the complete model of our proposed Affective Learning methodology. The following model (Figure 1) shows the complete learning process where the affective computing module is connected with traditional E-Learning system 1.0 through the evaluation module. Our new contribution having the fusion and emotion recognition module makes the whole system more adequate and advanced. Figure 1 shows the complete Affective E-Learning model where we can receive learner emotions with four different attributes (Facial, Speech, Gesture, and Bio-physical Signals). We have considered two emotion-attributes viz. Facial Images captured through spot detection technique and Bio-physical Signals captured using Heart Rate, Skin Conductance and Blood Volume Pressure sensing techniques. Images are captured through QHM495LM-3207 web camera placed in the front of the learner and ProComp5 machine is used to sense Heart Rate, Skin Conductance and Blood Volume Pressure. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.