IBM Watson AI Travel Assistant

A comprehensive travel recommendation system built with IBM Watson AI that provides real-time streaming responses for travel planning queries.

🌟 Overview

This project demonstrates how to integrate with IBM Watson AI to create an intelligent travel assistant that can help with:

Destination recommendations
Itinerary planning
Travel tips and advice
Local recommendations
Budget-friendly suggestions

🏗️ Architecture

Core Components

Authentication Layer: IBM Cloud IAM OAuth 2.0 integration
Streaming API Handler: Real-time response processing using Server-Sent Events (SSE)
Error Management: Comprehensive error handling for production use
Modular Design: Reusable functions for easy integration

Watson AI Integration

Deployment: Uses IBM Watson AI streaming endpoint
Region: US-South (us-south.ml.cloud.ibm.com)
Protocol: HTTPS with Bearer token authentication
Response Format: Server-Sent Events (SSE) with JSON chunks

📁 Project Structure

Travel_Agent_Ai/
├── travel_ai_notebook.ipynb    # Complete Jupyter notebook implementation
├── arsh.env                  # Your API key file (create this)
└── README.md                 # This file

File Descriptions:

travel_ai_notebook.ipynb: Interactive notebook with step-by-step implementation and detailed explanations
arsh.env: Your API key configuration file (you need to create this)

🚀 Quick Start

Prerequisites

Python 3.7+
IBM Cloud account with Watson AI service
Required Python packages: requests, re, json

Installation

Clone or download the project files

git clone https://github.com/arshdeepyadavofficial/Travel_Agent_Ai
cd Travel_Agent_Ai

Install dependencies
```
pip install requests python-dotenv
```
Set up your IBM Cloud API key

Create a .env file in the project root directory and add your IBM Cloud API key:
```
API_KEY=your_actual_ibm_cloud_api_key_here
```
How to get your IBM Cloud API key:
1. Go to IBM Cloud Console
2. Navigate to Manage → Access (IAM) → API Keys
3. Click Create to generate a new API key
4. Copy the key and paste it in your .env file
Important Notes:
- Never share your API key publicly
- Add .env to your .gitignore file
- The key should be without quotes or extra spaces

Usage

Option 1: Run the Jupyter Notebook (Recommended)

Set up your API key (see Configuration section above)

Start Jupyter Notebook

jupyter notebook travel_ai_notebook.ipynb

Run the cells in order - the notebook includes detailed explanations for each step

Option 2: Run as Python Script (if available)

# Make sure your .env file is set up first
python travel_ai_notebook.py  # If you convert the notebook to .py

🔧 Configuration

Getting Your IBM Cloud API Key

Sign up/Login to IBM Cloud
- Visit IBM Cloud
- Create an account or login
Create Watson AI Service
- Go to Catalog → AI/Machine Learning → Watson Studio
- Create a service instance
Generate API Key
- Navigate to Manage → Access (IAM) → API Keys
- Click Create button
- Give it a name (e.g., "Travel AI Assistant")
- Copy the generated API key
Create Environment File
- Create a .env file in your project directory
- Add your API key:
```
API_KEY=your_actual_ibm_cloud_api_key_here
```

Alternative Configuration Methods

Method 1: Custom Environment File (arsh.env)

API_KEY=your_ibm_cloud_api_key_here

Method 2: System Environment Variables

export API_KEY="your_ibm_cloud_api_key_here"

API Configuration Example

# The code automatically loads from .env or arsh.env files
# Your .env file should contain:
API_KEY=abcd1234-efgh-5678-ijkl-9012mnopqrst

# The deployment URL is already configured in the code

💡 Features

✅ Core Functionality

Real-time Streaming: See AI responses as they're generated
Authentication: Secure IBM Cloud IAM integration
Error Handling: Comprehensive error management
Modular Design: Easy to integrate and extend

✅ Travel Assistant Capabilities

Destination planning and recommendations
Itinerary creation (multi-day trips)
Travel tips and packing advice
Local recommendations (restaurants, attractions)
Budget-friendly travel options

✅ Technical Features

Server-Sent Events (SSE) parsing
JSON response chunk assembly
Token-based authentication
Production-ready error handling

📚 Usage Examples

Once your API key is configured, you can use the travel assistant for various queries:

Basic Travel Query

# Ask about destinations
travel_ai_assistant("I'm planning a trip to Paris. Can you suggest popular tourist attractions?")

Itinerary Planning

# Multi-day trip planning
travel_ai_assistant("Create a 3-day itinerary for Rome including must-see attractions")

Travel Tips

# Packing and preparation advice
travel_ai_assistant("What should I pack for a winter trip to Iceland?")

Local Recommendations

# Local experiences and food
travel_ai_assistant("Best authentic street food in Bangkok?")

🔐 Authentication Flow

API Key: Your IBM Cloud API key
Token Request: POST to IBM IAM endpoint
Access Token: Temporary token (expires in 3600 seconds)
Bearer Auth: Token used in Watson AI requests

# Authentication process
token_response = requests.post(IAM_URL, data={
    "apikey": API_KEY,
    "grant_type": 'urn:ibm:params:oauth:grant-type:apikey'
})
access_token = token_response.json()["access_token"]

📡 API Response Format

Watson AI returns streaming responses in SSE format:

id: 1
event: message
data: {"choices": [{"index": 0, "delta": {"role": "assistant", "content": "Hello"}}]}

id: 2
event: message  
data: {"choices": [{"index": 0, "delta": {"role": "assistant", "content": " there"}}]}

The code parses these chunks and assembles the complete response.

🛠️ Technical Details

Streaming Response Parsing

# Extract data lines from SSE format
data_lines = re.findall(r'data: (.+)', response_text)

# Parse each JSON chunk
for line in data_lines:
    data = json.loads(line)
    content = data['choices'][0]['delta'].get('content', '')
    full_response += content

Error Handling

error_messages = {
    400: "Bad Request - Check your message format",
    401: "Unauthorized - Verify your API key", 
    403: "Forbidden - Check your permissions",
    404: "Not Found - Verify deployment URL",
    429: "Rate Limited - Wait before making more requests",
    500: "Internal Server Error - Try again later",
    503: "Service Unavailable - Watson AI may be down"
}

🧪 Testing

Test Categories Covered

Destination Planning
- Family vacation recommendations
- Beach and food destinations
- Seasonal travel timing
Itinerary Creation
- Multi-day trip planning
- Budget-friendly options
- City-specific itineraries
Travel Tips
- Packing recommendations
- Family travel advice
- Solo travel safety
Local Recommendations
- Authentic dining experiences
- Hidden gems and local culture
- Cultural experiences

Running Tests

Execute the notebook cells in order, or run:

# Test basic functionality
response = travel_ai_assistant("Plan a trip to Tokyo")

# Test multiple query types
test_multiple_queries()

🚨 Error Codes & Troubleshooting

Code	Issue	Solution
401	Authentication failed	Check API key validity
404	Deployment not found	Verify deployment URL
429	Rate limited	Wait and retry
500	Server error	Retry after delay

Common Issues

"❌ API key is missing" Error
- Ensure your .env or arsh.env file exists in the project directory
- Check that the file contains: API_KEY=your_actual_key_here
- Make sure there are no extra spaces around the equals sign
- Verify the API key is valid and not expired
Authentication Failures
- Double-check your API key from IBM Cloud console
- Ensure the API key has proper permissions for Watson AI
- Try regenerating the API key if it's old
File Not Found Issues
- Make sure you're running the notebook from the correct directory
- Check that .env or arsh.env file is in the same folder as the notebook
- Use absolute paths if needed
Network Issues
- Check internet connectivity
- Verify firewall settings allow HTTPS requests
- Try different network if corporate firewall blocks requests
Response Parsing Issues
- Check if your Watson deployment supports streaming
- Verify API version compatibility
- Ensure deployment URL is correct

🔄 Production Considerations

Security

Store API keys in environment variables
Use secrets management systems
Implement request rate limiting
Add input validation and sanitization

Performance

Implement response caching
Add connection pooling
Use async requests for multiple queries
Monitor token expiration

Monitoring

Log API usage and errors
Track response times
Monitor token refresh cycles
Set up alerting for failures

🚀 Enhancement Ideas

Immediate Improvements

Add conversation history support
Implement token refresh mechanism
Add response caching
Create web interface (Flask/FastAPI)

Advanced Features

Integration Options

📄 License

This project is open source. Feel free to use, modify, and distribute.

🤝 Contributing

Fork the repository
Create a feature branch
Make your changes
Add tests if applicable
Submit a pull request

📞 Support

For issues and questions:

Check the troubleshooting section
Review IBM Watson AI documentation
Create an issue in the repository

�‍💻 Author

Arshdeep Yadav
BTech Computer Science Engineering

�🙏 Acknowledgments

IBM Watson AI team for the excellent API
IBM Cloud documentation and examples
Open source community for inspiration

Made with ❤️ for travelers worldwide 🌍✈️

Name		Name	Last commit message	Last commit date
Latest commit History 5 Commits
.gitignore		.gitignore
README.md		README.md
travel_ai_notebook.ipynb		travel_ai_notebook.ipynb

arshdeepyadavofficial/Travel_Agent_Ai

Folders and files

Latest commit

History

Repository files navigation