As sophomores in high school, we feel stressed when we do not understand the course and it would be beneficial to have another teacher to assist us. Finding the right teacher for a student can be a tedious and difficult process.
What it does
Prodigy allows anyone to create hyper-personal classrooms with ease. Hyper-personal classrooms are environments where teachers can develop a bond with their students. This bond is crucial to helping a student succeed. When a student finds the right teacher, they learn so much more, and they become engaged. Teachers share their experiences, stories, and knowledge with their students and this makes the learning environment more enjoyable.
How we built it
Prodigy is designed to make that relationship between students and teachers virtual and more available to people worldwide. We developed the website to simplify the student's search process to find the correct teacher for them. We also developed the website to help teachers contact students around the world within minutes. With the power of machine learning and lots of code, we can enhance the platform's capabilities to make searching a course and recommending courses fast and easy.
Challenges we ran into
We had a problem connecting the frontend code to the backend code, but through hard work and perseverance, we were able to stabilize the connection. We also had an issue with validating the login/signup form, but we got ideas from one another and resolved the issue. We first tried to use user-user collaborative filtering, but quickly found out that this recommendation system was ineffective due to the lack of user data. We solved this problem by creating a content based recommendation system that uses natural language processing to compare course similarity, instead of comparing user data.
Accomplishments that we're proud of
We are proud to create a platform where teachers can create personalized classrooms. This platform can help students get instruction from anywhere.
What we learned
We learned a lot about reactjs, and developing frontend web applications. We also learned about Django and Flask, two amazing python frameworks essential for our backend development. We learned how to use Natural Language Processing to locate key words in the descriptions of our courses. We then learned how to compare the similarity of courses using a cosine similarity matrix. We also learned how to use the Twilio API to connect the video of users and allow them to screenshare in a room. We then learned how to use ngrok to make a public url for the video room.
What's next for Prodigy
Prodigy will go on a public domain and will be available to the public. We will also adjust the website if you are a student or a teacher, and it will be more personalized.