Course Matrix: Your AI-Powered Academic Planner

Overview

Course Matrix is an AI-powered platform designed to streamline the course selection and timetable creation process for undergraduate students at UTSC.

By taking user preferences into consideration, Course Matrix is able to leverage AI to intelligently provide personalized course recommendations and generate optimized timetables.

Motivation

At UTSC, students face challenges when it comes to managing their academic and personal schedules. The process of selecting the optimal courses and ensuring that course prerequisites are met can be time-consuming and overwhelming, especially for international and first-year students. The current systems for dealing with this problem are fragmented as they require students to check multiple websites. This leads to inefficiencies and potential errors in course selection.

Course Matrix addresses these problems by providing a centralized, user-friendly platform that integrates the course database with advanced AI features. Our goal is to help students reduce scheduling errors, save valuable time, and provide a flexible, personalized approach to course planning.

Installation

Prerequisites

You will need to install the following prerequisites to get started: NodeJS version 20.10.0 or above

Getting Started

Follow these steps to install this application

1. Clone this repo:

git clone https://github.com/UTSC-CSCC01-Software-Engineering-I/term-group-project-c01w25-project-course-matrix.git

2. Install npm packages

cd ./course-matrix
cd ./frontend
npm install
cd ../backend
npm install

3. Configure environment variables: Create a .env file in /backend and populate it with the following:

NODE_ENV="development"
PORT=8081
CLIENT_APP_URL="http://localhost:5173"
DATABASE_URL=[Insert Supabase Project URL]
DATABASE_KEY=[Insert Supabase Project API key]

OPENAI_API_KEY=[Insert OpenAI API Key]
PINECONE_API_KEY=[Insert Pinecone API Key]
PINECONE_INDEX_NAME="course-matrix"

BREVO_API_KEY=[Insert Brevo API Key]
SENDER_EMAIL="[email protected]"
SENDER_NAME="Course Matrix Notifications"

The DATABASE_URL variable should contain your Supabase project url and the DATABASE_KEY should contain your Supabase project’s API key. To learn how to create a new Supabase project: see here. Likewise, the OPENAI_API_KEY variable should contain your OpenAI Project API Key, the PINECONE_API_KEY should contain your Pinecone Project API Key, and the BREVO_API_KEY should contain your Brevo API Key. Note that for the purposes of this project, we will provide the grader with all necessary API keys and URLs.

4. Configure environment variables for frontend. Create a .env file in /frontend and populate it with the following:

VITE_SERVER_URL="http://localhost:8081"
VITE_PUBLIC_ASSISTANT_BASE_URL=[Insert vite public assistant bas URL]
VITE_ASSISTANT_UI_KEY=[Insert vite assistant UI key]

Running the Application

To run the application locally:

1. Run backend:

cd ./backend
npm run dev

2. Run frontend

cd ../frontend
npm run dev

Navigate to the url given by the environment variable CLIENT_APP_URL in .env to access the frontend web page. This should be http://localhost:5173 initially.

Tech Stack and Software Architecture Pattern

• Tech Stack: We utilize the PERN (PostgreSQL, Express, React, and Node) tech stack.
  • The frontend and backend codebases both use TypeScript as a common language, which provides type safety across both codebases.
  • Additionally, we chose to utilize Supabase as a cloud-managed wrapper for our PostgreSQL database. There are many benefits of this such as built-in authentication and authorization, managed hosting and scalability, as well as team based project management.
  • We chose to use a relational database instead of a NoSQL database due to the structured nature of our data models (Users, Courses, Offerings, Timetables).
  • For more info about our tech stack, refer to this Google Docs document.
• Software Architecture Pattern: 3-Tier Architecture.

Contribution

We use Jira as our ticketing website. Also, Course Matrix follows the git flow process. For more info, see the sections below for the Branch Naming Rules and the Contribution Steps.

Branch Naming Rules

• The main branch holds the production-ready code.
• The develop branch is the integration branch where new features and non-breaking fixes are added and tested.
• Feature branches and bugfix branches are created off the develop branch and are merged back into it when complete. Both feature branches and bugfix branches follow the naming convention {initials}/{jira-ticket-number}-{descriptive-title} (e.g. ax/scrum-2-add-login-ui could be an example of a feature branch name and ax/scrum-3-fix-unresponsive-login-button could be an example of a bugfix branch name).
• Release branches are created off the develop branch and are merged into the develop and main branches when complete. Release branches follow the naming convention release/{version-number} (e.g. release/1.0).
• Hotfix branches are created off the main branch and are merged into the develop and main branches when complete. Hotfix branches follow the naming convention hotfix/{version-number} (e.g. hotfix/1.0.1).

Contribution Steps

1. Once you start working on a Jira ticket, set the status of that ticket to In Progress and create the associated branch for that ticket on GitHub.

2. Commit and push the necessary changes for your branch.

3. Create a pull request (PR) for your branch:

   • For hotfix or release branches, create one PR to merge it into the main branch and another one to merge it into the develop branch.
   • For feature branches, create one PR to merge it into the develop branch.

4. Get a PR approval from at least one other team member.

5. Merge the PR and mark the associated Jira ticket with a status of Done.

   • Just as a FYI, our GitHub repo is configured to strictly use the Squash and merge option for PRs with the default commit message set as the PR title.