An Effective Model for Forecasting Travel Consumer Demand Using Big Data Analysis

As a result of gathering information from multiple consumer centers, big data (BD) assists in analyzing traveler patterns and developing a unique marketing plan tailored to the target demographic. BD tourism forecasting is a relatively new academic field because of the challenges in capturing, gathering, and modeling this sort of data due to its inherent privacy and economic importance. The growth rate of cruise tourists has slowed down after years of rapid expansion. Investing in homeports, cruise ships, and promotional activities carries a growing danger of financial loss. To make investment decisions and prepare for the future, it is necessary to predict tourism demand. We present the least-squares vector regression (LSVR) model with the gravitational search method for forecasting demand for cruise tourism (FCT) based on BD to improve forecasting performance. As a part of the proposed model forecasting demand for cruise tourism based on big data (FDCT-BD), hyper-parameters of the LSVR model are improved using an algorithm and by comparing these models with various configuration combinations. This paper forecasts tourist arrivals based on internet BD from a search engine and online review platforms and the comparative advantage of multi-platform forecasting over single-platform forecasting based on online review data. However, the results show that the methodology’s recommended framework is successful and that BD may estimate cruise tourist demand with enhanced performance and accuracy 93.8% and 97.9%, respectively.