Scheduling college courses is frustrating. Our project solves this problem by using AI-powered course scheduling to generate the best class schedule based on your completed courses and preference.
Many students struggle with scheduling their courses efficiently due to complex requirements such as prerequisites, co-requisites, and personal time constraints.
The goal was to create a tool that helps students automatically generate their optimal course schedule without conflicts while considering their academic needs.
What it does
My AI Advisor suggests optimal course schedules based on:
Previously taken classes
Academic requirements such as prerequisites and co-requisites.
The system:
Uses Claude 3 Sonnet AI to process the data.
Integrates a knowledge base with course paths and timing data.
Ensures that no classes conflict with each other.
How we built it
Tech Stack:
Backend:
Python
AI Model:
Claude 3 Sonnet
Cloud Services:
AWS (Buckets for storage, Knowledge Base for data)
Frontend:
HTML/CSS (Flask Template)
Database:
A student model to store user inputs like classes taken, constraints, and personal information.
Key Components:
Flask App:
Collects user inputs (name, classes taken, time constraints) via forms.
Sends data to the cloud (AWS) for processing.
Claude 3 Sonnet AI:
Generates personalized course suggestions based on the user's data.
Knowledge Base:
Stores data on course requirements, timing, and class availability.
AWS Integration:
Stores and retrieves dynamic data that informs the AI’s recommendations.
Challenges we ran into
Complexity of Scheduling:
Handling multiple dependencies like prerequisites and co-requisites for courses, and ensuring that all classes fit into available timeslots.
Integrating Multiple Technologies:
Debugging the interaction between Flask, AWS, and Claude 3 Sonnet was challenging due to different technologies involved.
Data Consistency:
Ensuring that the data fed into the AI model was formatted right.
Accomplishments that we're proud of
AI Integration:
Successfully integrated Claude 3 Sonnet AI to generate personalized, conflict-free class schedules.
User-Friendly Interface:
Built a clean, functional Flask app that takes user input and provides course scheduling recommendations.
Scalable Data Model:
Created a flexible student model to store and manage user data, making the system easy to scale.
AWS Knowledge Base:
Set up an AWS-based knowledge base to store course paths, requirements, and timing data for easy retrieval and processing.
What we learned
Flask and Cloud Integration:
Gained valuable experience in using Flask for building web applications and integrating them with cloud services like AWS.
AI in Education:
Learned how to use AI for practical applications like course scheduling, where real-time, dynamic data plays a crucial role.
Data Management:
Understood the complexities of managing and feeding data into an AI system for personalized recommendations.
Problem-Solving:
Learned how to approach and solve complex scheduling problems that require multiple layers of logic and data analysis.
What's next for My AI Advisor
Improved Scheduling Algorithm:
Work on refining the scheduling algorithm to optimize for factors like workload balance, class proximity, and more personalized suggestions.
Advanced Features:
Implement features such as recommending electives or factoring in past student performance for even better recommendations.
User Feedback:
Incorporate user feedback to further improve the accuracy of the class suggestions and overall user experience.
Additional Data Sources:
Integrate more data sources (e.g., peer reviews, student grades) to refine the AI’s recommendations.
Log in or sign up for Devpost to join the conversation.