ProphetBnB: Airbnb Price Prediction & Host Segmentation

Welcome to ProphetBnB, a comprehensive Python project for predicting Airbnb listing prices and segmenting hosts for actionable insights. ProphetBnB combines robust machine learning, interactive visualizations, and a Streamlit web app for seamless exploration of Airbnb data.

---

Table of Contents

• Features
• Project Structure
• Installation
• Usage
• Dataset
• Dependencies
• Contributing
• License

---

Features

🏷️ Price Prediction

• Predicts Airbnb listing prices using regression models
• Flags underpriced and overpriced listings

👥 Host Clustering

• Segments hosts using clustering on price, reviews, and availability
• Provides insights for targeted marketing and strategy

📊 Interactive Visualizations

• Folium maps for geographic analysis
• Plotly charts for cluster and feature exploration

🚀 Streamlit App

• Intuitive interface for data exploration and prediction
• Accessible locally or via web deployment

---

Project Structure

AirBnB-PriceSense/
│
├── data/
│   ├── raw/               # Original datasets (CSV/JSON)
│   └── processed/         # Cleaned datasets for modeling
│
├── notebooks/
│   ├── 01_EDA.ipynb       # Exploratory Data Analysis
│   └── 02_Modeling.ipynb  # Regression & Clustering Models
│
├── src/
│   ├── data_preprocessing.py   # Data cleaning & feature engineering
│   ├── model_training.py       # ML model training & evaluation
│   └── visualizations.py       # Plotting & mapping functions
│
├── app/
│   └── prophetbnb_app.py       # Streamlit web app
│
├── assets/
│   └── logo.png                # Logo/images
│
├── environment.yml             # Conda environment
├── requirements.txt            # Pip dependencies
└── README.md                   # Project overview

---

Installation

1. Using Conda (Recommended)

# Navigate to the project folder
cd path/to/AirBnB-PriceSense

# Create environment
conda env create -f environment.yml

# Activate environment
conda activate ProphetBnB

2. Using Pip / Virtualenv

# Create virtual environment
python -m venv ProphetBnB_env

# Activate environment (Windows)
ProphetBnB_env\Scripts\activate

# Activate environment (macOS/Linux)
source ProphetBnB_env/bin/activate

# Install dependencies
pip install -r requirements.txt

---

Usage

1. Run Jupyter Notebooks

jupyter notebook

• notebooks/01_EDA.ipynb: Explore dataset, visualize distributions
• notebooks/02_Modeling.ipynb: Train regression & clustering models

2. Run Streamlit App

streamlit run app/prophetbnb_app.py

• Explore predictions and visualizations interactively
• Supports local and web deployment

---

Dataset

• Place your Airbnb dataset in data/raw/listings.csv
• The script src/data_preprocessing.py creates a cleaned dataset in data/processed/listings_clean.csv

Recommended Dataset Sources:

• Inside Airbnb
• Any Airbnb dataset you have access to

---

Dependencies

• Python 3.10+
• pandas, numpy, scikit-learn
• matplotlib, seaborn, plotly
• geopandas, folium, streamlit, streamlit-folium
• joblib, pyyaml
• Jupyter Notebook

See environment.yml & requirements.txt for full details.

---

Contributing

1. Fork the repository & create a new branch
2. Make your changes & test thoroughly
3. Submit a pull request with a clear description

---

License

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

---

Example: Streamlit App

How to Begin:

1. Choose your data source from Airbnb, CSV, or web
2. Set your filters for price, guests, reviews, etc.
3. Click Analyze Listings for insights

Supported Data:

• InsideAirbnb, CSV (with at least: id, name, price)
• For best experience: use clean, recent datasets

---

Empower your Airbnb decisions with smart predictions and interactive insights!