This project - AI-powered itinerary planner - is our submission for the Smart India Hackathon '24, designed to create personalized travel plans for users. It dynamically collects user inputs through a chatbot interface, processes the data to generate a tailored itinerary, and allows for real-time modifications based on user preferences, travel conditions, and past behavior. The backend uses Django, while the front-end consists of vanilla HTML, CSS, and JavaScript.
- Project Overview
- Features
- User Interaction Flow
- Technical Stack
- Installation
- Usage
- Data Structure
- Saving User Data for Continual Learning
- Contributors
This project enables users to create personalized travel itineraries. The system collects user information (e.g., number of travelers, destinations, vacation dates, activity preferences), and uses an AI-driven engine to suggest the best itinerary. Users can modify their itinerary based on their preferences in a chatbot-style interface. The system continually learns from the user’s travel patterns and provides better suggestions for future trips.
- Dynamic Chatbot Interaction: Collects user details such as number of travelers, destinations, travel dates, and preferences.
- Real-time Itinerary Generation: Uses AI to create custom itineraries based on user input, weather data, and nearby must-visit spots.
- Continual Learning: Saves and analyzes user data over time to offer personalized recommendations in future trips.
- User Behavior Analysis: Learns travel patterns like preferred activities, budgets, and destinations.
- Django Backend: Handles user data and itinerary generation requests.
- Frontend with Vanilla JavaScript: Offers a seamless user experience with chatbot-style interactions.
- Welcome Message: The chatbot welcomes the user and asks how many people are traveling and for basic trip details (e.g., destinations, vacation type).
- Form Submission: The user submits travel details (dates, location, preferences) through dynamic forms.
- Itinerary Display: The AI generates an itinerary in an easy-to-read table format and allows the user to modify it.
- Continual Feedback Loop: Users can modify the itinerary by interacting with the chatbot, and their feedback is incorporated.
- Finalization: The user confirms the finalized itinerary, and the system saves the user’s preferences for future trips.
- Backend: Django (Python)
- Frontend: HTML, CSS, JavaScript
- AI/ML: OpenAI GPT-4 API for itinerary generation
- Database: Django ORM for managing user data
- Media Storage: Django media directory for storing user behavior data
-
Clone the repository:
git clone https://github.com/shravan-18/itinerary-planner.git cd itinerary-planner -
Install dependencies:
pip install -r requirements.txt
-
Set up Django migrations:
python manage.py makemigrations python manage.py migrate
-
Run the Django server:
python manage.py runserver
- Access the Web App: Open your browser and go to
http://localhost:8000/. - Interact with the Chatbot: Provide trip details, such as the number of travelers, destination, and dates.
- Get Your Itinerary: The system generates an itinerary based on your inputs.
- Modify Itinerary: Use the chatbot interface to customize the itinerary further.
- Finalize Itinerary: Confirm your trip plan, which is saved for future analysis.
The user data is saved constantly after each itinerary to track behavioral patterns and improve future suggestions. The data is stored in a structured format using Django’s media directory, enabling continual learning through RAG (Retrieval-Augmented Generation).
We welcome contributions! Feel free to submit issues or pull requests.