Forecasting short-term tourism demand with a decomposition-ensemble strategy

A novel model, SSA-SVR, effectively forecasts short-term tourism demand fluctuations in the Mt. Siguniang Scenic Area of China's Sichuan Giant Panda Sanctuaries, addressing critical gaps in existing research. Unlike traditional approaches that focus on medium to long-term predictions, this study leverages a decomposition-ensemble strategy combined with machine learning algorithms to capture high volatility, nonlinearity, and instability in daily passenger flow data. The model demonstrates promising accuracy, offering valuable insights for government policymakers, scenic area managers, and tourists alike. This advancement could significantly improve resource allocation, visitor management, and policy planning in UNESCO World Heritage Sites facing similar challenges.

Feng, Y., Hao, J., Sun, X., & Li, J. (2022). Forecasting short-term tourism demand with a decomposition-ensemble strategy. Procedia Computer Science, 199, 879–884. https://doi.org/10.1016/j.procs.2022.01.110