Authentication using facial attributes obtained by convolutional neural networks in learning management system
Online assessments are increasingly present. This is even more evident with the great demand resulting from the 2020 pandemic. However, verifying the authenticity of the appraised is a task that still lacks new solutions. In this sense, this work presents two plugins for Moodle, a Learning Management System widely used in different levels of education. The first, called Face Verification Quiz, serves to register and verify the user's face in Moodle in a quiz activity with this feature enabled. The second, called Face Verification Report, is to show a report of successes in this activity. Trained deep convolutional neural networks were used to quickly detect the contour of the face through the pixels of the eyes, nose and mouth. Another network was used for face verification, extracting a 128-dimensional vector from each face using the ResNet-34 neural network architecture. These plugins have been tested with 32 users. The results show the feasibility of use, with an average approval rate of 76% through a questionnaire applied to 16 users in optional online activities carried out in Moodle in late 2020 and early 2021. Despite the challenges caused by the COVID-19 pandemic, the solutions showed to be useful after this isolation period.