Title

Promoting Active Learning and Protecting Student Privacy for Smart Education

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Computer Science and Info Sys

Date of Award

Spring 2023

Abstract

The research on Smart Education is rapidly developing in transforming from traditional face-to-face learning to online teaching and remote learning. Smart Education provides e-learning classes to improve the personalized educational system that creates to recognize and assesses each student's unique learning styles, educational needs, and academic objectives. So, to increase the active learning experience of a student in a video tutoring class, interaction session is quite helpful for the teacher to assess the student’s performance which will improve their teaching standards to deliver the content efficiently. For the above reasons, most of the smart classes suggest learners turn on their webcams during a class which helps easily to detect whether the students are engaged or not by their facial expressions, head pose, and eye gazing to promote active learning. However, it exposes the student’s face to raise security concerns such as privacy invasion, misidentification, data breaches, and lack of regulation by the respective authorities either school or corporate. So, this study thus proposes a novel framework benefiting two systems, such as facial identity protection utilizing face de-identification (i.e., generating a fake face) and detecting students’ engagement through facial behavior. This proposed method consists of a Face de-identifier having attributes along with a cloaking technique to protect and preserve the facial privacy of a student, Facial properties generator to create a new dataset for student engagement having face expressions, face orientation, eye gaze, face landmarks, and face action units and a (Multilayer Perceptron) MLP based engagement detection model to calculate the percentage of the level of student engagement in a smart education.

Advisor

Yuehua Wang

Subject Categories

Computer Sciences | Physical Sciences and Mathematics

Link to Full Text
COinS