Hotel Booking Cancellation Prediction

Developed a machine learning model to predict hotel booking cancellations using customer and reservation data. The project involved data preprocessing, exploratory data analysis (EDA), feature engineering, handling class imbalance, and comparing multiple classification algorithms.

Technologies

Python

Pandas

NumPy

Matplotlib

Seaborn

Scikit-learn

XGBoost

Joblib

Key Features

Cleaned and preprocessed missing values and invalid records.

Performed comprehensive EDA with visualizations.

Applied One-Hot Encoding and feature scaling.

Compared Logistic Regression, Decision Tree, Random Forest, SVM, and XGBoost.

Evaluated models using Accuracy, Precision, Recall, F1-score, and ROC-AUC.

Saved the final trained model using Joblib.

Results

Improved prediction performance by comparing multiple algorithms.

Identified the most important factors influencing booking cancellations.