AI Food Weather-Based Recommendation System

This project is a web-based application that provides food recommendations based on weather conditions, user preferences (vegetarian, non-vegetarian, or any), and cuisine type. Additionally, the app suggests popular restaurants known for specific food items in the user's locality. The backend is built with Flask and integrates AI-based recommendation logic, while the frontend is developed using React.

---

Features

1. Weather-Based Food Recommendations: Fetches weather data for a given city and suggests food items suited to the current weather.
2. User Preferences: Allows users to specify dietary preferences (vegetarian, non-vegetarian, or any) and cuisine type.
3. AI-Enhanced Recommendations: Filters and refines recommendations using trained models and AI-based logic.
4. Restaurant Suggestions: Displays a list of well-known restaurants for the recommended food items in the user's locality.
5. Responsive Frontend: Provides a clean and interactive UI for users to input their preferences and view recommendations.

---

Technologies Used

Backend

• Flask: A lightweight Python web framework.
• scikit-learn: For AI-based filtering and recommendation logic.
• Google Generative AI API: To enhance food suggestions based on weather and user inputs.
• OpenWeatherMap API: For fetching real-time weather data.
• Google Places API: To find popular restaurants for the recommended food items in the user's locality.
• Flask-CORS: To handle cross-origin requests.
• FuzzyWuzzy: For fuzzy matching of inputs (e.g., city names, cuisine types).

Frontend

• React: For building a responsive user interface.
• Axios: For making API calls to the Flask backend.
• CSS: Custom styles for a clean UI.

---

Application Structure

project-root/
│
├── backend/
│   ├── ai_recommendations.py
│   ├── app.py
│   ├── config.py
│   ├── cuisines.json
│   ├── myFood.json
│   ├── recommendations.py
│   ├── requirements.txt
│   └── .env
│
├── frontend/
│   ├── public/
│   │   └── index.html
│   ├── src/
│   │   ├── components/
│   │   │   ├── FoodRecommendation.jsx
│   │   │   └── FoodRecom.css
│   │   ├── App.css
│   │   ├── App.js
│   │   ├── index.css
│   │   └── index.js
│   └── package.json
│
└── README.md

---

Prerequisites

1. Python 3.8+
2. Node.js 14+
3. npm or Yarn
4. API Keys for:
   • OpenWeatherMap API
   • Google Generative AI API
   • Google Places API

---

Installation Guide

Backend Setup

1. Navigate to the backend directory:

   cd backend
   python -m venv venv
   venv\Scripts\activate
   

2. Install Python dependencies:

   pip install -r requirements.txt
   pip install --upgrade Flask Werkzeug
   pip install flask-cors
   pip install python-Levenshtein
   

3. Set up the .env file: Create a .env file in the backend directory with the following keys:

   WEATHER_API_KEY=<Your_OpenWeatherMap_API_Key>
   GENAI_API_KEY=<Your_Generative_AI_API_Key>
   PLACES_API_KEY=<Your_Google_Places_API_Key>
   CORS_ORIGIN=http://localhost:3000
   

4. Run the Flask application:

   python app.py
   

   The backend will run on http://127.0.0.1:5000.

Frontend Setup

1. Navigate to the frontend directory:

   cd frontend
   

2. Install dependencies:

   npm i
   

3. Run the React application:

   npm start
   

   The frontend will run on http://localhost:3000.

---

Usage

1. Open the browser and navigate to http://localhost:3000.
2. Enter a city name and specify your preferences:
   • Dietary preference (vegetarian, non-vegetarian, or any).
   • Cuisine type (e.g., Indian, Continental, etc.).
3. Click on the "Get Recommendations" button.
4. View the list of recommended food items and popular restaurants in your locality for those items.

---

Dependencies

Backend

• flask
• flask-cors
• requests
• scikit-learn
• google-generativeai
• google-places
• python-dotenv
• fuzzywuzzy

Install all dependencies via:

pip install -r requirements.txt

Frontend

• react
• axios

Install all dependencies via:

npm install

---

Example Workflow

1. Input:
   • City: New York
   • Preference: Veg
   • Cuisine Type: Indian
2. Backend Process:
   • Fetch weather data for New York.
   • Use the weather and user inputs to request food suggestions from the Generative AI API.
   • Filter and refine results using the trained AI model.
   • Use Google Places API to find popular restaurants for the recommended food items in New York.
3. Output:
   • A list of vegetarian Indian food items suitable for the current weather in New York.
   • Suggested restaurants in New York known for these food items.

---

AI and NLP Techniques Employed

1. TF-IDF Vectorization

Transforms textual food data into numerical representations for efficient analysis.

2. Cosine Similarity

Measures similarity between user preferences and available food recommendations.

3. Fuzzy String Matching

Handles imperfect user inputs for better robustness.

---

Contributing

Contributions are welcome! Feel free to open issues or submit pull requests.

---