Population Monitoring and Forecasting Model Based on XG-Boost
<p>Shihan Ma<sup>1</sup>, Qiang Li<sup>1,2</sup></p> · Academic Journal of Computing & Information Science · 2024
The expansion of urban areas and the rise in population mobility have led to significant alterations in population structure and urban configuration, presenting substantial problems for monitoring population dynamics. Due of its immediacy and reliability, big data, including mobile phone signalling and geographic position, offers possibility for population dynamic monitoring and precise management. The study investigated the population size and age structure in Hebei province utilizing large data from Hebei Unicom mobile phone signalling and Baidu Huiyan, employing complex network analysis to create and examine population migration. Based on data from the sixth and seventh population censuses in Hebei Province, a monitoring model for population size, birth rate, and transient population has been established, employing the population development equation as the foundational model and the XG-Boost machine learning algorithm for identification. The model is ultimately adjusted by cellular signalling data and validated using census data. The result shows that the population growth rate is slowing down, aging intensification, and a spatial distribution of population movement characterized by a "multi-core" structure in Hebei Province. And the population monitoring model developed by cellular phone signals and other large datasets demonstrates practicality and accuracy, efficiently predicting population size and structure.