Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Examining Monitoring System: Detecting Abnormal Behavior In Online Examinations

Dinh An Ngo, Thanh Dat Nguyen, Thi Le Chi Dang, Huy Hoan Le, Ton Bao Ho, Vo Thanh Khang Nguyen, Truong Thanh Hung Nguyen

TL;DR

The methodology and the effectiveness of the system in mitigating the widespread problem of cheating in online exams are outlined and high accuracy and speed in detecting cheating in real-time scenarios are demonstrated.

Abstract

Cheating in online exams has become a prevalent issue over the past decade, especially during the COVID-19 pandemic. To address this issue of academic dishonesty, our "Exam Monitoring System: Detecting Abnormal Behavior in Online Examinations" is designed to assist proctors in identifying unusual student behavior. Our system demonstrates high accuracy and speed in detecting cheating in real-time scenarios, providing valuable information, and aiding proctors in decision-making. This article outlines our methodology and the effectiveness of our system in mitigating the widespread problem of cheating in online exams.

Examining Monitoring System: Detecting Abnormal Behavior In Online Examinations

TL;DR

The methodology and the effectiveness of the system in mitigating the widespread problem of cheating in online exams are outlined and high accuracy and speed in detecting cheating in real-time scenarios are demonstrated.

Abstract

Cheating in online exams has become a prevalent issue over the past decade, especially during the COVID-19 pandemic. To address this issue of academic dishonesty, our "Exam Monitoring System: Detecting Abnormal Behavior in Online Examinations" is designed to assist proctors in identifying unusual student behavior. Our system demonstrates high accuracy and speed in detecting cheating in real-time scenarios, providing valuable information, and aiding proctors in decision-making. This article outlines our methodology and the effectiveness of our system in mitigating the widespread problem of cheating in online exams.
Paper Structure (13 sections, 5 figures, 2 tables)

This paper contains 13 sections, 5 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. System
  4. Functionalities
  5. Abnormal Behavior Detection Tool
  6. Model Development
  7. Dataset Development
  8. Data Collection
  9. Data Preprocessing
  10. Model Experiment
  11. Model Evaluation
  12. Results
  13. Conclusion And Future Work

Figures (5)

  • Figure 1: System Overview: Before entering the exam room, individuals authenticate their identity as a student or proctor. The system aids proctors by monitoring student behavior, issuing warnings, and capturing images of rule breaches. If a student breaks the rules more than three times, the quiz locks. The proctor then decides to either unlock or end the exam.
  • Figure 2: The student’s screenshot during the exam is displayed. The left side shows a photo of the violation, with the system issuing a warning to the students. The right side shows a photo taken when the proctor resumes the exam after a suspension due to the student committing more than three violations.
  • Figure 3: Sample data of Normal (left) and Abnormal (right) class.
  • Figure 4: The pipeline follows a sequence of steps from data collection to model training. Data is collected from the laptop camera, split, and resized. Facial landmarks are extracted using Mediapipe for initial preprocessing. Images with all zero values are removed from the training set. The Euclidean distance formula is used to calculate distances between 19 selected landmark points. The final dataset is then used for training and validation.
  • Figure 5: Model architecture comparison between CNN and Mediapipe. We use Mediapipe to extract features instead of letting the model autonomously learn them, facilitating the learning process and reducing external environmental influences.