Skip to content

Latest commit

 

History

History
228 lines (168 loc) · 6.56 KB

README.md

File metadata and controls

228 lines (168 loc) · 6.56 KB

Spicescape 🍲

Discover the hidden flavors of your city with Spicescape, a one-of-a-kind app that unearths street food stalls not found on conventional maps. Whether you're a foodie or an adventurer, Spicescape is your trusted companion for finding, exploring, and enjoying hidden culinary gems.

Key Features 🚀

  1. Hidden Gem Locator
    Pinpoints unlisted street food stalls and vendors around you.

  2. Crowdsourced Recommendations
    Access reviews and ratings submitted by food lovers just like you.

  3. Personalized Suggestions
    Get tailored recommendations based on your location, preferences, and history.

  4. Interactive Map
    Explore food stalls through a sleek and intuitive map interface.

  5. Real-Time Updates
    Stay informed about operational timings, menu changes, and live events.

How Spicescape Works ⚙️

Spicescape uses advanced algorithms to deliver accurate, real-time recommendations and mapping services:

Core Algorithms and Formulas ⚙️

1. Haversine Formula 🌍

The Haversine formula calculates the great-circle distance between two points on a sphere, enabling accurate proximity filtering for food stalls.

Formula:

$$
d = 2r \cdot \arcsin\left(\sqrt{\sin^2\left(\frac{\phi_2 - \phi_1}{2}\right) + \cos(\phi_1) \cdot \cos(\phi_2) \cdot \sin^2\left(\frac{\lambda_2 - \lambda_1}{2}\right)}\right)
$$
  • ( d ): Distance between two points (in kilometers or miles)
  • ( r ): Radius of the Earth (approx. 6371 km)
  • ( \phi_1, \phi_2 ): Latitudes of the two points (in radians)
  • ( \lambda_1, \lambda_2 ): Longitudes of the two points (in radians)

Use Case in Spicescape:
This formula helps in filtering nearby food stalls based on user location.

2. Bounding Box Filtering 📦

Bounding box filtering is used to create a rectangular geographic area for quick search results by minimizing the Haversine calculations.

Formulas:
For the minimum and maximum latitude:

$$
\text{min\_lat} = \phi - \frac{\text{radius}}{r}
$$
$$
\text{max\_lat} = \phi + \frac{\text{radius}}{r}
$$

For the minimum and maximum longitude:

$$
\text{min\_lon} = \lambda - \arcsin\left(\frac{\text{radius}}{r \cdot \cos(\phi)}\right)
$$
$$
\text{max\_lon} = \lambda + \arcsin\left(\frac{\text{radius}}{r \cdot \cos(\phi)}\right)
$$
  • ( \phi, \lambda ): Latitude and longitude of the center point (in radians)
  • ( \text{radius} ): Search radius (in kilometers or miles)
  • ( r ): Radius of the Earth (approx. 6371 km)

Use Case in Spicescape:
Bounding box filtering narrows down the search area before applying precise calculations like Haversine.

3. Tokenization (Text Preprocessing for Search) ✂️

Tokenization breaks text input (e.g., food stall names or reviews) into smaller components for better search and NLP processing.

Tokenization Process:

  1. Convert text to lowercase: 
    input_text.lower()
  2. Remove punctuation: 
    re.sub(r'[^\w\s]', '', input_text)
  3. Split text into tokens: 
    tokens = input_text.split()

Example:
For input “Best hot dogs in town!”:

  • Step 1: Convert to lowercase: "best hot dogs in town"
  • Step 2: Remove punctuation: "best hot dogs in town"
  • Step 3: Split into tokens: ["best", "hot", "dogs", "in", "town"]

Use Case in Spicescape:
Tokenization is used in search functionalities, enabling fast and accurate food stall lookups based on keywords or reviews.

Example Integrations 🔗

  • Haversine Distance: Quickly find nearby food stalls using precise distance calculations.
  • Bounding Box Filtering: Narrow down search areas to improve performance for large datasets.
  • Tokenization: Enhance search accuracy for queries like "best biryani near me."

Crowdsourced Data Aggregation

Users contribute directly to Spicescape by submitting stall details, reviews, and photos. These are processed using a blend of natural language processing (NLP) and sentiment analysis to provide meaningful recommendations.

Getting Started 🛠️

Backend Setup

Follow these steps to get the Spicescape backend running locally:

1️⃣ Clone the Repository

Clone the project to your local system and navigate to the /backend directory:

git clone <repository_url>
cd spicescape/backend

2️⃣ Create a .env File

Inside the /backend folder, create a file named .env and add the following variables:

POSTGRES_DB=spicescape_db
POSTGRES_USER=spicescape_user
POSTGRES_PASSWORD=spicescape_password
SECRET_KEY=mysecretkey123

3️⃣ Build and Start the Backend

Run the following command:

docker-compose up --build

4️⃣ Access the Backend

Once the backend is running, you can access it at:

http://localhost:8000/api/food-stalls/

Example Query Parameters:

  • Filter by cuisine: 
    http://localhost:8000/api/food-stalls/?cuisine=italian
  • Filter by location and radius: 
    http://localhost:8000/api/food-stalls/?latitude=17.3850&longitude=78.4867&radius=2

Backend Commands:

  • Populate food stall data: 
    python manage.py populate_food_stalls data.csv
  • Retrieve nearby stalls: 
    python manage.py retrieve_nearby_stalls 17.3850 78.4867 5

Frontend Setup

The frontend is designed with React for a smooth, user-friendly experience. Here’s how to set it up:

1️⃣ Install Dependencies

Navigate to the /frontend folder and run:

npm install

2️⃣ Start the Application

Use the following command to build and run the frontend:

docker-compose up --build

Visit http://localhost:3000 in your browser.

Future Enhancements 🌟

  1. AI-Driven Recommendations
    Implement advanced machine learning models for even more accurate suggestions.

  2. Enhanced UI/UX
    Introduce modern design patterns and animations for a richer user experience.

  3. Gamification Features
    Add badges, achievements, and leaderboards to incentivize exploration.

  4. Vendor Integration
    Allow vendors to directly update their profiles and share promotions with users.

Contributing 🤝

We welcome contributions from the community! If you’d like to help improve Spicescape, feel free to fork the repository, make your changes, and submit a pull request.

License 📄

Spicescape is licensed under the MIT License. See the LICENSE file for details.