Title Hotel Booking Cancellation Prediction Using Deep Learning
Text
ClassificationDeveloped a binary classification pipeline to predict hotel reservation cancellations using a structured dataset of over 36,000 real-world booking records. The project involved cleaning and preprocessing raw tabular data, handling missing values, encoding categorical variables, and scaling numerical features using Pandas and Scikit-learn to ensure data quality before model training. Built and trained a deep learning model using the Keras Sequential API with TensorFlow, applying Dropout and EarlyStopping to prevent overfitting. Model quality was rigorously evaluated using confusion matrix, ROC curve, and classification metrics which able to achieving 84.5% accuracy and an AUC score of 0.92 on unseen test data, confirming strong generalization and label accuracy across both classes.