An Automated Anti-Cheat Framework for Online Exams in Pakistan

Abstract

In the past few decades E-learning in higher education is increased and play vital role in pandemic like COVID-19. Particularly, online examinations are conducted on e-learning platforms that leads to save time and other resources for both institutes and students. Also, where online examination has advantages but have impact of dis honesty of students in online examination. For this reason, numerous research is available that proposes methodologies and techniques for seamless execution of online examination and prevent the cheating in online examination. It is very important to have adequate solution for above problem. However, this study provides the comparative analysis of online examination techniques, tools performed on 50 studies from last five years 2017 to 2021. All the studies identified the lack of framework for automated online examination system for less developed countries. Furthermore, frameworks proposed, and the industrial or commercial tool do not provide the cost-effective, offline monitoring solution with less requirement of internet bandwidth which is very important for less developed countries to adopt the OEP systems. Hence there is a sheer need of developing an anti-cheat framework for less developed countries that predicts anti-cheating behavior of students as well as cost effective solution. Keeping this in view, an open-source framework for anti-cheat online examination in Pakistan has proposed that ensures improved ease of use of software and cheating detecting during offline monitoring and low network bandwidth. The framework also automated the cheating prediction reports. The proposed framework used NLP technique to find out the cheating prediction on student’s monitoring process and data. The benefit of using the NLP technique to generate the high accurate results as per studies. Now to make the cost-effective solution the framework is developed using cloud technology that also help to monitor the offline processes or activities and ultimately provide the usability of minimum internet bandwidth requirement. The framework not only easy to use for students but also provide the improved usability for instructors are administrators who are authorized to generate the automated reports. Furthermore, we validate the framework by conducting the exam for two different subjects from 11 students. Our framework successfully analyzed the cheating and non-cheating cases by achieving the 90% accuracy.