.

ISSN 2063-5346
For urgent queries please contact : +918130348310

Student’s Emotion Recognition through Facial Expressions during E-Learning using Fuzzy Logic and CNN classification

Main Article Content

Ati Jain, Dr. Hare Ram Sah, and Harsha Atre
» doi: 10.48047/ecb/2023.12.Si6.053

Abstract

COVID19 brought up with lots of changes in education system. In E-learning, classes and trainings were done on platforms like Zoom, Google meet etc. Identification of state of the mind of students those were attending sessions becomes difficult. Emotion recognition is one of the hot topics these days in fields like medical, business, education, academic research and many more. Many applications like cognitive computing, affective computing, computer vision, entertainment is widely used with emotion recognition and are at high demand. Technique like facial expressions recognition with identification of emotions like Anger, Disgust, Sad, Happy, Surprise, fear and Neutral can be judged to better understand. The proposed solution calculates concentration index of student and also give feedback about delivery of the class to teachers by student’s attention during the class. Implementation takes off traditional feedback method and comes up with original results on recognizing attention from student’s expressions. This is implemented with deep learning model like Convolutional Neural Network using Keras (python language) where built model will be checked through live data and FER2013 datasets for the emotion recognition. Also, by using concepts of Fuzzy Logic, Fuzzy rule sets are prepared and membership functions implemented by Mamdani MATLAB Software. Finally, accuracy of the model will be calculated and results will be compared. Such applications are useful for any online learning student’s/trainees/mentees that shows involvement, interest and attention of participants. Teaching learning process is improved with the help of such applications.

Article Details