Exam Proctoring Classification Using Eye Gaze Detection

Currently, as there is a spike in the number of examinations being conducted online, there is also an increasing need for proctors, a resource that soon got overwhelmed. Proctors are supposed to monitor each and every examinee individually to ensure that the code of conduct is being followed by the examinee, contrary to the offline examination system in which a large group of students are overlooked by a handful of invigilators. But even then, the rate of unfair means being opted by the examinees in online examinations is quite higher in comparison to offline examinations. Thus, making the current process extensive, tiresome and wasteful. To make the process more coherent we came up with Eagle Eye. This system would be implemented and made efficient with the use of the technique of Eye gaze detection using Artificial Intelligence and Machine Learning. At the beginning, of the examination, the examinee would be required to undergo a calibration test to determine the situations around the examinee and to set a designated border-box area into which the examinee would be required to get their eyeball movements tested. This data gathered from the detection would be used to examine whether or not the examinee is looking outside or inside the box, which in turn would be used to classify the data as fraudulent and/or fair respectively. If detected to be fraudulent by the system, an alert will be sent to the examiner and the examinee both so that further actions can be taken according to the mandate. To achieve the best prediction results, we have made our own dataset with the help of volunteers so as to obtain unfiltered and more authentic samples to work with. Impact Statement — Whilst the shift of the education sector into the online mode has been rather swift than not. Nevertheless, there are a few areas in which some modifications or improvements could be made to improve efficiency and decrease the need for enormous human capital. One of such areas is Online Examinations, to be specific, the proctoring method. Even with the presence of individual proctors with every single examinee, the rate of unfair means being opted by the examinees in online examinations is quite higher in comparison to offline examinations wherein a single proctor keeps a track of a group of examinees. Thus, making the current process extensive, tiresome and wasteful. To make the process more coherent we came up with Eagle Eye. A system which would use less financial and human resources and capital investment.