Booth Course Assistant Chatbot

A comprehensive chatbot designed to help Booth School of Business students with course-related queries, including course information, bidding history, and degree requirements.

Features

The chatbot provides assistance with:

• Course information and prerequisites
• Degree requirements
• Concentration requirements
• Course bidding history and analysis
• Course syllabi information
• Real-time course scheduling information

Example Usage

Here are some example questions you can ask the assistant:

1. Degree Requirements

• "What courses can I take to fulfill the Decisions requirement?"
• "What are the core course requirements?"

2. Concentration Requirements

• "Does 30131 fulfill requirements for Accounting concentration?"
• "What courses count towards Finance concentration?"

3. Course Information

• "When is investments offered next?"
• "What courses are offered on Monday evenings in Spring 2025?"

4. Bidding History

• "What are the bid points for 34106?"
• "How many points do I need for Investments?"

5. Course Syllabi

• "What are the prerequisites for Advanced Investments?"
• "What is the grading policy for Negotiations?"
• "What do we learn in 34106?"

6. Course Number to Title

• "What is the title for 35150?"

Backend Architecture

Core Components

• Framework: Flask-based REST API
• AI Model: GPT-4 powered by LangChain
• Memory: Conversation buffer for context retention
• Thinking Process: Transparent reasoning with step-by-step thought process

Tools and Integrations

1. Course Information Tools

   • Vector-based course search
   • Context-based course search
   • Course number to title mapping

2. Academic Requirements

   • Degree requirements checker
   • Concentration requirements validator

3. Bidding Analysis

   • Historical bidding data processor
   • Bid point analysis and recommendations

4. Data Processing

   • Excel to CSV conversion for bidding data
   • Course data vectorization
   • Syllabus parsing and analysis

Frontend Interface

Technologies

• Framework: Streamlit
• Features:
  • Clean, intuitive chat interface
  • Real-time response display
  • Expandable thinking process view
  • Chat history management
  • Raw response data access
  • Error handling with user feedback

User Interface Components

• Query input field
• Chat history display
• Thinking process expansion
• Response formatting
• Error notifications
• Clear chat functionality

Setup and Installation

Prerequisites

# Required Python version
Python 3.12 (Tested and recommended)

# Required environment variables
OPENAI_API_KEY=your_api_key_here

Installation Steps

1. Clone the repository:

git clone [email protected]:bilalbaqar/booth-chatbot.git

# If you dont want to use ssh to clone
# git clone https://github.com/bilalbaqar/booth-chatbot.git

cd booth-chatbot

2. Set up Python environment (choose one option):

Option A: Using Conda (Recommended)

# Create and activate conda environment
conda create --name env-booth-chatbot python=3.12
conda activate env-booth-chatbot

# Install dependencies
pip install -r requirements.txt

Option B: Using venv

# Create and activate virtual environment
python3.12 -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

3. Set up environment variables:

# Create .env file
touch .env  # On Windows: type nul > .env

# Add the following content to your .env file:
OPENAI_API_KEY=your_api_key_here

# You can do this via command line:
echo "OPENAI_API_KEY=your_api_key_here" > .env

# Or manually create/edit the .env file in your preferred text editor
# and add the following line:
# OPENAI_API_KEY=your_api_key_here

Note: Replace your_api_key_here with your actual OpenAI API key. You can get an API key from OpenAI's website.

Running the Application

1. Start the backend server:

python .\booth_agent\booth-agent.py --server --server

2. Start the frontend interface (in a new terminal):

streamlit run frontend\chatbot-frontend.py

The application will be available at:

• Frontend: http://localhost:8501
• Backend API: http://localhost:5000

Alternative Run Modes

The backend supports different modes of operation:

# Run in CLI mode for command-line interaction
python booth-agent.py --cli

# Run test queries
python booth-agent.py --test

# Show help
python booth-agent.py --help

Tools and Dependencies

Main Libraries

• langchain: For AI model integration and chain management
• openai: For GPT-4 API access
• flask: For backend API server
• streamlit: For frontend interface
• pandas: For data processing
• numpy: For numerical operations
• requests: For API communication

Data Processing Tools

• DocArrayInMemorySearch: For vector search capabilities
• CSVLoader: For data loading and processing
• OpenAIEmbeddings: For text embeddings