The volatility of tourism demand is often caused by some irregular events in recent years. Typically, inbound tourists are quite sensitive to various factors, including the exchange rate fluctuation, consumer price index, personal or household income or consumption expenditure. We combine these multivariate time series data onto an ingenious multi-factor fusion strategy to contribute to precise tourism demand forecasting. A novel hybrid deep learning forecasting approach is developed by integrating several modules such as improved complete ensemble empirical mode decomposition with adaptive noise, intrinsic mode functions classification, multi-factors fusion and predictors matching. The monthly tourist flow data of Shanghai inbounding from USA, Korea and Japan are conducted to verify the performance of the proposed approach, which outperforms all benchmark models for different prediction horizons. The experimental results show that introducing external influencing factors can improve the prediction accuracy significantly, and therefore confirm the rationality and validity of the proposed approach.
Accurate traffic flow forecasting is a fundamental component of Intelligent Transportation Systems and proactive urban mobility management. However, the inherent complexity of urban traffic flow, characterized by non-stationary dynamics and multi-scale temporal dependencies, poses significant modeling challenges. Existing spatio-temporal models often struggle to reconcile the discrepancy between static physical road constraints and highly dynamic, state-dependent spatial correlations, while their reliance on fixed temporal receptive fields limits the capacity to disentangle overlapping periodicities and stochastic fluctuations. To bridge these gaps, this study proposes a novel Multi-scale Graph-Decoupling Spatial–temporal Network (MS-GSTN). MS-GSTN leverages a Hierarchical Moving Average decomposition module to recursively partition raw traffic flow signals into constituent patterns across diverse temporal resolutions, ranging from systemic daily trends to high-frequency transients. Subsequently, a Tri-graph Spatio-temporal Fusion module synergistically models scale-specific dependencies by integrating an adaptive temporal graph, a static spatial graph, and a data-driven dynamic spatial graph within a unified architecture. Extensive experiments on four large-scale real-world benchmark datasets demonstrate that MS-GSTN consistently achieves superior forecasting accuracy compared to representative state-of-the-art models. Quantitatively, the proposed framework yields an overall reduction in Mean Absolute Error of up to 6.2% and maintains enhanced stability across multiple forecasting horizons. Visualization analysis further confirms that MS-GSTN effectively identifies scale-dependent spatial couplings, revealing that long-term traffic flow trends propagate through global network connectivity while short-term variations are governed by localized interactions.