COLLEGE ADMISSION TEST WITH INTELLIGENT PROCTORING

Abstract

The main objective of this study was to design and develop a College Admission Test With Intelligent Proctoring.  Eye recognition and face motion algorithm were developed to monitor acts of cheating.  Captured images captured were used for evaluation after the examination was conducted as a proof if cheating really was committed. This study is a web based proctoring that could detect acts of cheating, provide feedback for examinees who commit acts of cheating and provide a list of examinees who possibly cheat. The study used an OpenCV Open Source Computer Vision algorithm to monitor examinee’s eye and face motion while taking the examination. The images captured by the eye and face detection algorithm was evaluated and used as a proof if cheating was committed.The system was evaluated using ISO/IEC 9126. A Questionnaire duly validated by experts was employed for user’s to validate the system. Mean and Standard Deviation were used for statistical data analysis. Series of system testing were undertaken together with the users and experts.The F- Measure was used to determined the harmonic mean of precision and recall in terms of  the accuracy of the system. Based on the F-Measure, the study revealed that it has 84% accuracy and 92 % precision. Eventually, it showed that the result of the study was accurate and precise. The overall evaluation of the system by the users were very effective. Finally, the system was “very effective” based on the overall evaluation of IT Experts.