In this paper student’s emotions such as excitement, happiness, confusion, sadness, desire and surprise, will be analysed by an Emotion AI. Therefore the neural network model, designed to capture the facial expression, is used. Deep learning is the emerging techniques to process large datasets of images with Kera’s using TensorFlow backend. Convolution Neural Network is an artificial neural network that has specialization in detection and classification. Convolution neural network has hidden layers called convolution layers. This layer consists of neurons. Facial emotion recognition usually employs a training and testing stage to produce the desirable output. The emotion of the students plays the vital role to determine the student interest in attending classes. Facial expressions are among the most universal forms of body language. The facial expressions are almost similar throughout the world. The facial expression, movement of head, eye, mouth helps to identify the emotions of the students so that the level of interest of student can be predicted form the emotion analysis of students. For example: A smile can be used to indicate happiness. Facial expression reveals the true feelings about a situation. Then, after collecting those information, e-learning quality can be improved and enhanced. The reaction of the students is analyzed during the teaching and learning course. Thus, the mood of students can be predicted easily which help to improve the e - learning environment. The feedback will be provided to teachers to enhance the teaching and learning process in e-learning.
Table of Contents
INTRODUCTION
OBJECTIVES
RELATED WORK
METHODOLOGY
A) System Architecture
B) Algorithm
C) Image Preprocessing
D) Feature Engineering
E) Concentration level measurement
F) Design of Convolution Neural network
DATA COLLECTION
RESULT AND ANALYSIS
A) Face detection sample output
B) Facial key points identification
C) Concentration level
D) Emotion Classification
E) Analysis
CONCLUSION
REFERENCES
Objectives and Research Themes
The primary objective of this research is to improve the quality of e-learning during the COVID-19 pandemic by developing an emotion-aware AI system that detects and monitors student engagement, providing real-time feedback to teachers.
- Development of a convolution neural network for emotion recognition.
- Predicting student mood through facial expression and head movement analysis.
- Measuring student concentration levels in virtual classroom environments.
- Enhancing pedagogical effectiveness through automated feedback loops.
Excerpt from the Book
METHODOLOGY
System Architecture define the overall procedure and working of the emotion AI learning system. The main blocks in architecture are image preprocessing, feature engineering, concentration level metrics, convolution neural network, classification of emotions and feedback to teacher. From the video of student learning in online classes, the picture frames are collected at different time interval. Facial images are focused. Emotions and concentration level are classified at the final stage.
The emotion AI learning system consist of student’s facial image as input. Convolution Neural network is applied to the images using different hidden layers. Training dataset of images are used to train the images. Different concentration level is also measured such as high-level concentration, and low-level concentration. Then, emotions of student are obtained. The emotions may be excitement, happiness, confusion, sadness, desire and surprise. On the basis of these emotions and concentration level, the feedback is collected and provided to the teacher.
Summary of Chapters
INTRODUCTION: Provides context on the challenges of e-learning during the COVID-19 pandemic and the need for interactive tools to maintain student engagement.
OBJECTIVES: Outlines the specific goals, including designing neural network architectures and enhancing e-learning environments through emotion analysis.
RELATED WORK: Reviews historical and contemporary research on facial expression analysis and its application in educational technology.
METHODOLOGY: Details the technical approach, covering system architecture, algorithms, image preprocessing, and the design of the convolutional neural network.
DATA COLLECTION: Describes the acquisition of training data from Kaggle and the collection of real-world testing data from Pashchimanchal Campus.
RESULT AND ANALYSIS: Presents the empirical findings, including face detection outputs, key point identification, and emotion classification graphs.
CONCLUSION: Summarizes the effectiveness of the system in measuring student concentration and providing actionable feedback to educators.
Keywords
Artificial Intelligence, Neural network, e-learning, covid-19, emotion, University, Deep learning, Face detection, Facial expression, Concentration level, Emotion recognition, Educational technology, Student engagement, Computer vision, Feedback system.
Frequently Asked Questions
What is the core purpose of this research?
The research aims to enhance the quality of e-learning during the COVID-19 pandemic by using AI to monitor student emotions and concentration, ensuring teachers receive feedback to adjust their teaching methods.
What are the central thematic areas?
The work focuses on Artificial Intelligence, emotion recognition, convolutional neural networks, and their practical application in virtual classroom environments.
What is the primary research goal?
The goal is to design an AI-driven system capable of classifying student emotions and measuring concentration levels to improve student engagement in online learning.
Which scientific methods are utilized?
The study utilizes Deep Learning, specifically Convolutional Neural Networks (CNNs), combined with image processing techniques for facial key point detection and feature engineering.
What does the main body of the paper address?
It addresses the system architecture, the technical algorithm for emotion identification, and the empirical analysis of test data collected from university students.
Which keywords characterize the work?
Key terms include Artificial Intelligence, Neural network, e-learning, emotion, student engagement, and facial expression analysis.
How is the student's concentration level determined?
Concentration is measured by tracking head rotation and eye visibility; if a student is centered with eyes open, it indicates higher concentration, whereas deviations suggest low focus.
How does the system benefit teachers?
The system provides teachers with direct feedback on the emotional state and engagement levels of the class, allowing them to optimize the learning process in real-time.
- Arbeit zitieren
- Hari K. C. (Autor:in), 2020, Emotion Artificial Intelligence as improvement for e-Learning during COVID-19, München, GRIN Verlag, https://www.grin.com/document/933196
