Application of data mining for identifying and predicting room bookings - window of opportunity

Abstract

This study investigates the search patterns for predicting hotel bookings. Using Expedia’s search and purchase data, I identified the user’s booking window and which events have a higher effect on the booking likelihood. The tourism industry has seen exponential growth over the last two decades, much due to global socioeconomic changes, globalization and internet massification. Portugal is finally reaping its share of profit and continues winning “best destination” prizes year after year. It is now part of a very competitive ecosystem where distribution plays a determinant role in whether the touristic product survives or not. Big players like Expedia and Booking.com have taken control of a big chunk of the market’s revenue because they understood that the large amounts of data they have enabled predicting demand hence offering highly competitive deals. Only by understanding the booking drivers one can negotiate better distribution deals and lower commissions through making better sales predictions and enhancing marketing and revenue strategies, hence the purpose of this study being to use data mining to find patterns in room bookings, enabling this industry to become an even more important source for the country’s GDP. Through the analysis of consumer behavior and booking times and the use of time series analysis and machine learning, it is possible to find patterns that can be applied across not only Hospitality but other industries as well. By looking for the connections and the relevance of each feature on the final predictions, a new window of opportunity will open for marketing and sales professionals.