Leveraging Machine Learning to Analyze and Predict Cultural Trends in Global Tourism

Abstract

Cultural trends, economic factors, and digital transformations shape fast-evolving global tourism. Machine learning is used in this study to analyze and predict cultural shifts in tourism on the basis of the fusion model by combining the architectures of EfficientNet and DiCENet. The model also integrates diverse data sources, which include online travel reviews, social media content, and macroeconomic indicators, to create a detailed framework to explain new travel behavior. The results indicate a strong recovery in positive traveler sentiment post-pandemic, alongside significant growth in sustainable travel, digital nomadism, and wellness tourism. The major influence on tourism demand was the level of economic stability, specifically GDP per capita and unemployment rates. In comparison, machine learning baseline models showed an accuracy rate of 92.5% in detecting and predicting trends under the EfficientNet-DiCENet fusion model. These results provide a basis for stakeholders in the industry, including policymakers and businesses, to make decisions that take advantage of evolving passenger air travel preferences in order to align the strategies. Even though the study showcases the impact of machine learning in tourism analytics, a number of shortcomings, such as data bias and the unpredictability of external disturbances, have been identified as research areas for the future. In general, this work contributes to the use of AI-driven models for data-driven tourism forecasting and cultural trend analysis.