This project delves into the realm of cybersecurity with a focus on the Internet of Things (IoT), examining the potential vulnerabilities and attack vectors within IoT ecosystems. Through the utilization of cutting-edge large language model tools, this project aims to analyze and understand the evolving landscape of IoT attacks, contributing valuable insights to enhance cybersecurity defenses in IoT environments.

Our project consists in designing a website based on visioncapabilities to prevent cheating in online exams and monitor multiplestudents at once using advanced machine learning models.Indeed, the website offers image verification authentication so thatthe system allows only one login per user, so that the user cannot makeunfair means. In addition, the system will check the user's image at everymoment of login and also during the exam using facial recognitiontechnology. Then, the teacher can create an exam, view the exam history,share the exam details with the students, view the questions, update, deletequestions, view the live monitoring of the exam and view the students'monitoring logs. On the other hand, the student has access only to take anexam. In addition, the monitoring is done by creating logs : detection of cellphones, detection of more than 1 person at the exam, estimation of gaze:Estimation of student body position and eye movements and finallyrecorded images of students every 4 seconds.Finally, we store our monitoring data in the cloud for visualisation andanalysis using a dashboard.