Algorithm for Border Port Personnel Flow Analysis and Economic Vitality Assessment Based on Spatiotemporal Big Data

The conventional techniques utilizing statistical methods and single time series are relatively inadequate to analyze the complex spatiotemporal correlations and economic connections, leading to the challenges in performing refined dynamic analyses of population flow and its influence on the economy. Hence, This article will use a novel fusion approach that integrates the spatiotemporal big data analysis and graph neural network to build a unified model for analyzing the dynamic behavior and economic potential of population flow at border ports. From a methodological perspective, indi vidual mobility patterns are characterized via spatiotemporal alignment and trajectory reconstruction, where this article further create the spatiotemporal graph of flows to describe the interregional migration relations. After that, we design a spatiotemporal graph neural network along with a multi-scale dynamic modeling strategy to predict population flows. As indicated by the experiments, the proposed method performs exceptionally well in population flow prediction tasks, obtaining MSE of 0.071 and MAE of 0.174, which demonstrates significant outperformance compared to the conventional approach. Through this study, we prove the effectiveness of our method in achieving refined population flow analysis and providing reliable data and decision-making support for assessing regional economic vitality.