ML Driven Emotion Identification for Feedback Analysis in E-Learning Platforms

ML Driven Emotion Identification for Feedback Analysis in E-Learning Platforms

U.Venkataramana Reddy1,Y.Sanjeeva Reddy2,Y.Mani Baskar Reddy3

1U. Venkataramana Reddy,

venkataramanareddy.cse@gmail.com,

Dr. M.G.R Educational and Research and institute

2Y. Sanjeeva Reddy,

reddysanjeev304@gmail.com,

Dr. M.G.R Educational and Research and institute

3Y. Mani Baskar Reddy,

manibhaskarreddybasannagari@gmail.com,

Dr. M.G.R Educational and Research and institute

Dr. G. Soniya Priyatharshini

soniya.cse@drmgrdu.ac.in

Dr. M.G.R Educational and Research and institute

Mr. Saravanan Elumalai,

Saravanan.e@drmgrdu.ac.in

Dr. M.G.R Educational and Research and institute

Abstract:  Feedback analysis for E-Learning platforms using deep learning methods for facial expression analysis is proposed in this work. They suggests a nonintrusive technique to evaluate the psychological impacts of elearning platforms by considering feedback data of users in the form of videos and analyzes emotions from video and decides the user feedback level. This process automates the understanding of feedback in a short amount of time for anyone, without requiring specific knowledge of the system.Convolutional Neural Networks (CNNs), a category of deep learning, has become increasingly popular for its ability to precisely analyze and comprehend emotions from facial photos. These models may efficiently learn and extract essential features by being trained on large datasets of tagged facial expression photos. This application takes video feedback as input for various E-learning platforms and analyzes emotions and displays the predicted emotions on video. A publicly accessible dataset is used to test the proposed technique, and the results show that it achieves 90.3% accuracy. This indicates that the approach is highly effective in accurately recognizing facial emotions.

Keywords: Convolutional Neural Network (CNN) face images, Artificial Intelligence (AI), E learning, Machine learning, Deep learning.