Hyper-Parameter Tuning with Grid and Randomized Search Techniques for Predictive Models of Hotel Booking

The tourism and hospitality industries rely heavily on the hotel industry, making it crucial for hotels to optimize their business strategies. Predicting hotel bookings is one strategy that can enable hotel management to make informed decisions about pricing, marketing, and resource allocation. In this paper, we explore the use of machine learning algorithms to predict hotel bookings, focusing on two types of hotels, leisure and business hotels. We use a dataset from Kaggle containing information on hotel bookings, including 32 features such as hotel type, arrival date, departure date, and booking date. We pre-process the data by converting categorical variables into numerical variables, imputing missing values, and performing feature scaling. We evaluate various machine learning algorithms on the dataset, including Gradient Boosting Classifier, Random Forest Classifier, and Logistic Regression Classifier, using accuracy as the evaluation metric. Our results show that these three algorithms perform well on the given dataset, with the Gradient Boosting Classifier achieving the highest accuracy of 86.5%, followed by Random Forest Classifier with an accuracy of 85.2%, and Logistic Regression Classifier with an accuracy of 80.7%. Our findings demonstrate the potential of machine learning techniques in predicting hotel bookings and can be useful to hotel management in optimizing their business strategies.