Predicting Hotel Booking Cancellations Using Machine Learning

Project Overview

INN Hotels Group faces significant revenue loss due to booking cancellations. This project aims to develop a machine learning solution to predict cancellation likelihood, enabling proactive measures to mitigate losses.

Objective

Develop a machine learning model to forecast booking cancellation probabilities, allowing for:

1. Dynamic pricing strategies: Adjust room rates based on predicted cancellation probabilities.
2. Targeted retention campaigns: Identify high-risk bookings for personalized retention efforts.
3. Optimized overbooking policies: Manage reservations more effectively to minimize revenue loss due to cancellations.

By accurately predicting cancellations, INN Hotels Group can minimize inventory loss and maximize revenue.

---

Table of Contents

1. Installation
2. Usage
3. Data
4. Notebooks

---

Installation

To run the project locally, follow these steps:

1. Clone the repository:

   git clone https://github.com/kachiann/Hotel_Booking_Cancellations.git

2. Navigate to the project directory:

   cd Hotel_Booking_Cancellations
   

3. Install dependencies:

   pip install -r requirements.txt

---

Usage

pip install gradio

2. python gradio_app.py

---

Data

The data folder contains all datasets used in this project

---

Methodology

1. Data Preprocessing: Data cleaning, feature engineering, and splitting.
2. Model Training: Building and training the machine learning model.
3. Evaluation: Assessing model performance using various metrics.
4. Deployment: Using Gradio for model deployment and interactive testing.

---

Notebooks

The notebooks folder contains Jupyter notebooks for different stages of the project:

• EDA_Booking_Cancellations.ipynb: Exploratory Data Analysis for understanding data patterns and features.
• Train_model_Booking_Cancellations.ipynb: Training the model
• Deployment.ipynb: Deploying it using Gradio for interactive demonstrations.

---