Project: Enhancing Online Learning with Eye Tracking and AI-Generated Quizzes
Inspiration
The inspiration for our project stems from the challenges faced by students like my little brother who struggled with staying focused during online learning sessions post-COVID-19. Despite the convenience of online education, distractions on the internet often led to a decline in his academic performance. Recognizing that many students faced similar issues, we felt compelled to find a solution to help them stay engaged and focused during online classes.
What We Learned
Throughout the development of our project, we learned a great deal about the intersection of technology and education. Specifically, we delved into the complexities of implementing real-time eye tracking and leveraging AI for generating relevant quizzes. We gained insights into the challenges faced by students in the online learning environment and the potential of technology to address these challenges.
How We Built Our Project
We built our solution by combining innovative technologies to create a comprehensive platform for online education. At the core of our solution is real-time eye tracking, which monitors students' iris movements to detect if they are distracted or looking away from the screen. When distraction is detected, alerts are sent to students to intervene promptly.
In addition to eye tracking, we integrated AI to generate quizzes tailored to the content being taught. This involved extracting keywords from the lesson material and dynamically generating multiple-choice questions based on these keywords. Teachers can customize the quizzes according to their preferences, ensuring alignment with the lesson objectives.
For the frontend development, we utilized getstream.io's WebRTC capabilities to facilitate seamless live meetings between teachers and students. We opted for Next.js for load optimization, ensuring optimal performance even during peak usage times.
Challenges Faced
While developing our project, we encountered several challenges that required creative problem-solving:
Implementing WebRTC: Integrating WebRTC for real-time communication posed initial difficulties, especially in ensuring compatibility across different browsers and devices. However, through thorough testing and troubleshooting, we were able to overcome these challenges and achieve reliable live meeting functionality.
Generating MCQs with AI: One of the major challenges we faced was devising a method to generate multiple-choice questions (MCQs) that were relevant and coherent. Specifically, mapping extracted keywords from the lesson material to form meaningful sentences proved to be a daunting task. However, by refining our AI algorithms and leveraging natural language processing techniques, we successfully tackled this challenge.
Despite these challenges, our determination to enhance online learning experiences for students propelled us forward, ultimately leading to the successful development of our solution.
By addressing the issue of distraction and offering personalized quizzes, we believe our project has the potential to significantly improve the efficacy of online education and empower students to achieve their academic goals more effectively.
Built With
- flashtext
- getstream.io
- mediapipe
- natural-language-processing
- nextjs
- nltk
- pke
- sense2vec
- t5-transformers
- tailwind
- webrtc
Log in or sign up for Devpost to join the conversation.