BearlyAI

💡 Inspiration

Our project was inspired by the challenges faced by students in navigating the extensive course offerings at Berkeley. We recognized a gap in the existing system where students often struggle to find courses aligned with their interests and career goals. Our aim was to leverage AI technology to simplify this process, making it more intuitive and tailored to individual student needs. By integrating advanced search capabilities and AI-driven recommendations, we aspire to enhance the academic experience for the Berkeley community.

💻 What it does

Our project is a web-based application that enables Berkeley students to discover courses that align with their academic interests and career aspirations. By inputting keywords, students can search for relevant courses. The platform uses AI to analyze course descriptions and recommend classes that match the student’s interests. It also connects to external resources like Rate My Professor, providing comprehensive information to help students make informed decisions.

⚙️ How it works

The application operates on a simple yet effective mechanism. The frontend consists of a user-friendly webpage with a textbox for inputting search keywords. Once a keyword is entered, the backend, powered by the Spacey LLM application, processes this input. It searches through a pre-clustered course catalog, identifying courses that match the keyword. The relevant course suggestions are then sent back to the webpage for the user's perusal.

🔨 How we built it

Our team employed a range of technologies to bring this project to life. We built the frontend using React, creating an interactive and responsive interface. The backend was developed in Flask, which handles requests and interacts with the Spacy API for course recommendations. We opted for a file-based system over a traditional database for storing course information, enhancing the simplicity and maintainability of our application.

🧠 Challenges we ran into

Achieving the right balance between accuracy and breadth of course recommendations required fine-tuning the AI algorithms. Additionally, designing a user interface that was both intuitive and feature-rich presented a significant challenge.

🏅 Accomplishments that we're proud of

We are particularly proud of successfully creating a tool that bridges the gap between student needs and academic offerings. Our application’s ability to provide personalized course recommendations based on AI analysis is a significant achievement. We're also proud of our teamwork and the collaborative effort that went into overcoming the technical challenges we faced.

📖 What we learned

Throughout this project, we gained invaluable experience in API integration, AI implementation, and user interface design. We learned the importance of understanding user needs and the challenges in creating a tool that addresses those needs effectively. This project also enhanced our skills in collaborative problem-solving and project management.

🚀 What's next

Moving forward, we plan to refine the AI algorithms to improve recommendation accuracy further. We also aim to incorporate more comprehensive data sources, including professor ratings and course feedback, to enrich the information provided to users. Additionally, we are considering expanding this platform to include other universities, making it a universal tool for academic course discovery.

Installing and Running

pip install spacy pip install nltk pip install flask python -m spacy download en_core_web_lg. Run the web app by running python3 backend.py in terminal, and interact with it by accessing localhost:5000 in a browser

Video Demo

Our video demo will walk through the key features of our application, showcasing the ease of use and efficiency of the course discovery process. We will demonstrate real-time searches, the breadth of information provided, and the seamless integration of various technologies that make this tool a valuable asset for the Berkeley student community.