Hotel Recommender System

A full-stack hotel recommendation system that leverages a React Native frontend with a voice agent and a Python backend. The backend processes user queries through a multi-step pipeline to return the most relevant hotel recommendations.

---

Table of Contents

• Challenge Summary
• Our Solution
• Prerequisites
• Setup Instructions
  • Backend
  • Frontend

---

Challenge Summary

Finding the perfect hotel can be challenging when travelers have specific preferences, such as pet-friendliness, free parking, or proximity to city attractions. The challenge is to build an AI-based hotel recommendation system that takes a user's natural language prompt describing their preferences and returns a sorted list of the most relevant hotels.

---

Our Solution

Our solution consists of:

• Frontend:
  A React Native application with a voice agent that allows users to speak their hotel preferences. The app sends the transcribed prompt to the backend and displays the recommended hotels.

• Backend:
  A Python API (FastAPI/Uvicorn) that processes hotel data and user prompts through a multi-step pipeline:

  1. Query Improvement: Optionally refines the user query.
  2. Validation: Checks if the request is valid and relevant.
  3. Unrestricted Request Handling: If the request is broad, returns the top 10 pre-ranked hotels.
  4. Relevant Feature Extraction: Identifies which hotel features are relevant to the query.
  5. Constraint Creation: Builds search constraints based on the query.
  6. Scoring: Calculates a relevance score for each hotel.
  7. Ranking: Returns the top 10 hotels ranked by score.

The backend pipeline is visualized below:

---

Prerequisites

• React (for frontend)
• npm (comes with React)
• Python 3.12 (for backend)
• pip (for Python dependencies)

---

Setup Instructions

1. Clone the Repository

git clone https://github.com/your-username/your-repo.git
cd your-repo

---

2. Backend

cd backend
python3.12 -m venv venv
source venv/bin/activate  # On Windows: .\venv\Scripts\activate
pip install -r requirements.txt
pre-commit install

To start the backend server:

uvicorn app:app --reload

---

3. Frontend

Open a new terminal window/tab, then:

cd frontend
npm install

To start the frontend development server:

npm run dev