Novel Hybrid Emotion Recognition Framework (NHERF) for Virtual Learning Assessment Model a Review.

Authors

  • Fayaz Ahmad Fayaz, Dr. Arun Malik

Abstract

Covid-19 has taught us how to learn from home via with the advancement in ICT, the concept of virtual learning and its applications like virtual classroom is a shared online space where the learners and the tutor work together simultaneously and participants take the advantage to attend lectures of delegates delivered irrespective to geographical boundaries conceptualizes the distributive computing. Usually, these interactions take place through e- learning tools. The participants have tools to present learning content in different formats, as well as to implement collaborative and individual activities. In this type of interaction, the teacher has the important role of the moderator who guides the learning process and supports group activities and discussions quizzes and other teaching learning activities. To make the system versatile and foolproof with feedback, the proposed Hybrid Human Emotion Recognition (HER) system has a role to play. Emotions play an important role in the learning process. Considering the learner's emotions is essential for distance mode of learning or Virtual electronic learning (e-learning) systems. Research has proposed that system should induce and conduct the learner's emotions to the suitable state. But, at first, the learner's emotions have to be recognized by the system. Like the context of HER. The emotions can be recognized by asking from the user, tracking implicit parameters, voice recognition (VR), facial expression recognition (FER), voice signals (VS) and gesture recognition (GR). While taking a lead system with Multi factor Hybrid emotion recognition framework combining face, gesture, voice signals, text, self-reporting characteristics and wearable sensor based human activity recognition can be implemented with comparison among other methods. It would be also noted that learner’s emotion detection is biased toward the learner’s age, demographic variables, geographic location & culture. Further system has to be trained by validity and reliability of training label to ensure by human experts hence work can be concentrated on designing appropriate deep learning and creating real databases for training the model. Hence the proposed system would be having its extensive applications in diversified fields.

Published

2020-02-29

Issue

Section

Articles