Method and system for regional population dynamic monitoring and prediction based on spatiotemporal big data

Abstract

This application provides a method and system for regional population dynamic monitoring and prediction based on spatiotemporal big data. The method includes: acquiring historical population characteristic data for a preset region, including monthly and annual historical population characteristic data; acquiring a preset prediction model set, filtering and processing it to obtain preset prediction model sets that meet prediction requirements, and marking them as a preset short-term prediction model set and a preset medium-to-long-term prediction model set, respectively; collecting real-time population characteristic data for the preset region; and combining it with the monthly and annual historical population characteristic data, processing it separately using the preset short-term and medium-to-long-term prediction model sets to obtain corresponding monthly and annual population prediction value sets; further processing to obtain the final monthly and annual population prediction values, thereby realizing the technology for regional population dynamic monitoring and prediction based on spatiotemporal big data.

Description

Technical Field

[0001] This application relates to the field of spatiotemporal big data technology, and more specifically, to a method and system for regional population dynamic monitoring and prediction based on spatiotemporal big data. Background Technology

[0002] With the acceleration of urbanization and the increasing frequency of regional population movement, accurate population dynamic monitoring and forecasting have become an important support for land spatial planning and public resource allocation. Traditional population monitoring relies on censuses and sampling surveys, which have shortcomings such as poor timeliness, low spatial resolution, and high labor costs, making it difficult to capture short-term fluctuations and long-term change patterns in the population.

[0003] Existing forecasting technologies mostly use a single model or single-scale data, failing to effectively distinguish between short-term and medium-to-long-term forecasting needs, and lacking deep integration of historical data and real-time spatiotemporal big data, which easily leads to forecasting bias. At the same time, the application of multi-source spatiotemporal big data suffers from problems such as insufficient data connection and unreasonable model selection, making it impossible to balance the accuracy of monthly and annual forecasts and failing to meet the needs of refined governance.

[0004] Therefore, there is an urgent need for a technology that integrates multi-scale historical data and real-time data, and achieves accurate short-term and medium-term predictions through reasonable model selection. This technology would address the pain points of existing methods, such as monitoring lag and insufficient prediction accuracy, and provide reliable technical support for regional population dynamic management. Summary of the Invention

[0005] The purpose of this application is to provide a method and system for regional population dynamic monitoring and prediction based on spatiotemporal big data. This method involves acquiring historical population characteristic data of a preset region, including monthly and annual historical population characteristic data; acquiring a preset prediction model set, filtering and processing it to obtain preset prediction model sets that meet the prediction requirements, and labeling them as preset short-term prediction model sets and preset medium-to-long-term prediction model sets, respectively; collecting real-time population characteristic data of the preset region; and combining the monthly and annual historical population characteristic data with the preset short-term and medium-to-long-term prediction model sets, respectively, to obtain corresponding monthly and annual population prediction value sets. The final monthly population prediction value is obtained by processing the monthly population prediction value set, and the final annual population prediction value is obtained by processing the annual population prediction value set, thus realizing the technology for regional population dynamic monitoring and prediction based on spatiotemporal big data.

[0006] This application also provides a method for regional population dynamics monitoring and prediction based on spatiotemporal big data, including the following steps: Obtain historical population characteristic data for the preset area, including monthly and annual historical population characteristic data; Obtain a preset prediction model set, perform filtering to obtain a preset prediction model set that meets the prediction requirements, and mark them as the preset short-term prediction model set and the preset medium- and long-term prediction model set respectively. Real-time population characteristic data of the preset area is collected, and combined with the monthly historical population characteristic data and the annual historical population characteristic data, it is processed by the preset short-term prediction model set and the preset medium- and long-term prediction model set respectively to obtain the corresponding monthly population prediction value set and annual population prediction value set. The final monthly population forecast value is obtained by processing the set of monthly population forecast values. The final annual population forecast is obtained by processing the set of annual population forecast values.

[0007] Optionally, in the regional population dynamic monitoring and prediction method based on spatiotemporal big data described in this application, the step of obtaining historical population characteristic data of a preset area, including monthly historical population characteristic data and annual historical population characteristic data, includes: Obtain historical population characteristic data for the preset area, including monthly and annual historical population characteristic data; The monthly historical demographic data includes monthly resident population statistics, monthly natural population growth rate, monthly mechanical population growth rate, and monthly registered regional resident population. The annual historical demographic data includes annual resident population statistics, annual natural population growth rate, annual mechanical growth rate, annual regional population density, annual regional GDP, and annual regional employment rate.

[0008] Optionally, in the regional population dynamics monitoring and prediction method based on spatiotemporal big data described in this application, the step of obtaining a preset prediction model set, performing screening processing to obtain a preset prediction model set that meets the prediction requirements, and marking them respectively as a preset short-term prediction model set and a preset medium- and long-term prediction model set, includes: Obtain the preset prediction model set, including the preset exponential smoothing model set, the preset ARIMA model set, the preset Prophet model set, the preset spatial lag model set, and the preset XGBoost model set; Obtain the average absolute percentage error corresponding to each preset prediction model in the preset prediction model set; Obtain the first preset average absolute percentage error threshold and the second preset average absolute percentage error threshold; The average absolute percentage error corresponding to the preset exponential smoothing model set, the preset ARIMA model set, and the preset Prophet model set is compared with the first preset average absolute percentage error threshold to obtain a first comparison result. Based on the first comparison results, a first set of preset prediction models that meet the prediction requirements is obtained and marked as a set of preset short-term prediction models; The average absolute percentage error corresponding to the preset ARIMA model set, preset Prophet model set, preset spatial lag model set and preset XGBoost model set are compared with the second preset average absolute percentage error threshold to obtain the second comparison result. Based on the second comparison results, a second preset prediction model set that meets the prediction requirements is obtained and marked as the preset medium- and long-term prediction model set.

[0009] Optionally, in the regional population dynamic monitoring and prediction method based on spatiotemporal big data described in this application, the step of collecting real-time population characteristic data of the preset region, and processing it respectively with the monthly historical population characteristic data and the annual historical population characteristic data through the preset short-term prediction model set and the preset medium- and long-term prediction model set to obtain the corresponding monthly population prediction value set and annual population prediction value set, includes: Collect real-time population characteristic data of the preset area, including real-time population statistics, real-time birth rate, real-time death rate, real-time population inflow and real-time population outflow; The real-time population characteristic data and the monthly historical population characteristic data are processed by the preset short-term prediction model set to obtain the corresponding monthly population prediction value set. The real-time population characteristic data and the annual historical population characteristic data are processed by the preset medium- and long-term prediction model set to obtain the corresponding annual population prediction value set.

[0010] Optionally, in the regional population dynamics monitoring and prediction method based on spatiotemporal big data described in this application, the step of processing the monthly population prediction set to obtain the final monthly population prediction value includes: Statistical processing is performed on the monthly population forecast set to obtain the mean and standard deviation of the population forecasts. A preset reasonable prediction threshold range is obtained based on the mean and standard deviation of the predicted population values. The third comparison result is obtained by comparing each monthly population forecast value in the monthly population forecast set with the preset reasonable forecast threshold range. Based on the third comparison result, a set of reasonable values ​​for monthly population forecasts is obtained; The final monthly population forecast value is obtained by weighting the set of reasonable monthly population forecast values.

[0011] Optionally, in the regional population dynamics monitoring and prediction method based on spatiotemporal big data described in this application, the step of processing the annual population prediction set to obtain the final annual population prediction value includes: The preset residual threshold range is obtained based on the preset long-term trend residual method; The fourth comparison result is obtained by comparing the annual population forecast values ​​of each year in the set of annual population forecast values ​​with the preset residual threshold interval. Based on the fourth comparison result, a set of reasonable values ​​for annual population forecasts is obtained; The final annual population forecast is obtained by weighting the set of reasonable annual population forecast values.

[0012] Secondly, this application provides a regional population dynamics monitoring and prediction system based on spatiotemporal big data. The system includes a memory and a processor. The memory includes a program for a regional population dynamics monitoring and prediction method based on spatiotemporal big data. When the program for the regional population dynamics monitoring and prediction method based on spatiotemporal big data is executed by the processor, it performs the following steps: Obtain historical population characteristic data for the preset area, including monthly and annual historical population characteristic data; Obtain a preset prediction model set, perform filtering to obtain a preset prediction model set that meets the prediction requirements, and mark them as the preset short-term prediction model set and the preset medium- and long-term prediction model set respectively. Real-time population characteristic data of the preset area is collected, and combined with the monthly historical population characteristic data and the annual historical population characteristic data, it is processed by the preset short-term prediction model set and the preset medium- and long-term prediction model set respectively to obtain the corresponding monthly population prediction value set and annual population prediction value set. The final monthly population forecast value is obtained by processing the set of monthly population forecast values. The final annual population forecast is obtained by processing the set of annual population forecast values.

[0013] Optionally, in the regional population dynamics monitoring and prediction system based on spatiotemporal big data described in this application, the step of acquiring historical population characteristic data of a preset area, including monthly historical population characteristic data and annual historical population characteristic data, includes: Obtain historical population characteristic data for the preset area, including monthly and annual historical population characteristic data; The monthly historical demographic data includes monthly resident population statistics, monthly natural population growth rate, monthly mechanical population growth rate, and monthly registered regional resident population. The annual historical demographic data includes annual resident population statistics, annual natural population growth rate, annual mechanical growth rate, annual regional population density, annual regional GDP, and annual regional employment rate.

[0014] Optionally, in the regional population dynamics monitoring and prediction system based on spatiotemporal big data described in this application, the step of obtaining a preset prediction model set, performing screening processing to obtain a preset prediction model set that meets the prediction requirements, and marking them respectively as a preset short-term prediction model set and a preset medium- and long-term prediction model set, includes: Obtain the preset prediction model set, including the preset exponential smoothing model set, the preset ARIMA model set, the preset Prophet model set, the preset spatial lag model set, and the preset XGBoost model set; Obtain the average absolute percentage error corresponding to each preset prediction model in the preset prediction model set; Obtain the first preset average absolute percentage error threshold and the second preset average absolute percentage error threshold; The average absolute percentage error corresponding to the preset exponential smoothing model set, the preset ARIMA model set, and the preset Prophet model set is compared with the first preset average absolute percentage error threshold to obtain a first comparison result. Based on the first comparison results, a first set of preset prediction models that meet the prediction requirements is obtained and marked as a set of preset short-term prediction models; The average absolute percentage error corresponding to the preset ARIMA model set, preset Prophet model set, preset spatial lag model set and preset XGBoost model set are compared with the second preset average absolute percentage error threshold to obtain the second comparison result. Based on the second comparison results, a second preset prediction model set that meets the prediction requirements is obtained and marked as the preset medium- and long-term prediction model set.

[0015] Optionally, in the regional population dynamic monitoring and prediction system based on spatiotemporal big data described in this application, the step of collecting real-time population characteristic data of the preset region, and processing it in conjunction with the monthly historical population characteristic data and the annual historical population characteristic data through the preset short-term prediction model set and the preset medium- and long-term prediction model set respectively, to obtain the corresponding monthly population prediction value set and annual population prediction value set, includes: Collect real-time population characteristic data of the preset area, including real-time population statistics, real-time birth rate, real-time death rate, real-time population inflow and real-time population outflow; The real-time population characteristic data and the monthly historical population characteristic data are processed by the preset short-term prediction model set to obtain the corresponding monthly population prediction value set. The real-time population characteristic data and the annual historical population characteristic data are processed by the preset medium- and long-term prediction model set to obtain the corresponding annual population prediction value set.

[0016] As can be seen from the above, the regional population dynamic monitoring and prediction method and system based on spatiotemporal big data provided in this application acquires historical population characteristic data of a preset region, including monthly and annual historical population characteristic data; acquires a preset prediction model set, performs filtering and processing to obtain a preset prediction model set that meets the prediction requirements, and marks it as a preset short-term prediction model set and a preset medium- and long-term prediction model set, respectively; collects real-time population characteristic data of the preset region; and processes the monthly and annual historical population characteristic data respectively through the preset short-term and preset medium- and long-term prediction model sets to obtain corresponding monthly and annual population prediction value sets. The monthly population prediction value set is processed to obtain the final monthly population prediction value, and the annual population prediction value set is processed to obtain the final annual population prediction value, thereby realizing the technology of regional population dynamic monitoring and prediction based on spatiotemporal big data.

[0017] Other features and advantages of this application will be set forth in the following description and will be apparent in part from the description or may be learned by practicing embodiments of this application. The objectives and other advantages of this application may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A flowchart of a regional population dynamics monitoring and prediction method based on spatiotemporal big data provided in an embodiment of this application; Figure 2 This is a schematic diagram of a regional population dynamics monitoring and prediction method based on spatiotemporal big data provided in an embodiment of this application. Detailed Implementation

[0020] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0021] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0022] Please refer to Figure 1 , Figure 1 This is a flowchart of a regional population dynamics monitoring and prediction method based on spatiotemporal big data according to some embodiments of this application. This method is used in terminal devices, such as computers and mobile terminals. The method includes the following steps: S11. Obtain historical population characteristic data for the preset area, including monthly historical population characteristic data and annual historical population characteristic data; S12. Obtain the preset prediction model set, perform filtering processing to obtain the preset prediction model set that meets the prediction requirements, and mark them as the preset short-term prediction model set and the preset medium- and long-term prediction model set respectively. S13. Collect real-time population characteristic data of the preset area, and process it by combining the monthly historical population characteristic data and the annual historical population characteristic data with the preset short-term prediction model set and the preset medium- and long-term prediction model set respectively to obtain the corresponding monthly population prediction value set and annual population prediction value set. S14. Process the monthly population forecast set to obtain the final monthly population forecast value; S15. Process the annual population forecast set to obtain the final annual population forecast value.

[0023] It is important to note that traditional population monitoring relies on censuses and sampling surveys, which suffer from poor timeliness, low spatial resolution, and high labor costs. Existing forecasting technologies often employ single models or single-scale data, failing to differentiate between short-term and medium-to-long-term forecasting needs, and lacking deep integration of historical data with real-time spatiotemporal big data. This leads to forecasting biases and an inability to balance monthly and annual forecast accuracy. To address these issues, the first step is to acquire historical population characteristic data for the predefined region, including monthly and annual historical population characteristic data. Monthly historical population characteristic data includes monthly resident population statistics, monthly natural population growth rate, monthly mechanical population growth rate, and monthly regional permanent resident registration numbers. Annual historical population characteristic data includes annual resident population statistics, annual natural population growth rate, annual mechanical population growth rate, annual regional population density, annual regional GDP, and annual regional employment data. The system acquires a set of preset prediction models, including preset exponential smoothing models, preset ARIMA models, preset Prophet models, preset spatial lag models, and preset XGBoost models. These models are then filtered to obtain the preset prediction models that meet the prediction requirements, and are labeled as preset short-term prediction models and preset medium-to-long-term prediction models, respectively. Next, real-time population characteristic data for a preset region is collected and combined with monthly historical population characteristic data, processed using the preset short-term prediction model set to obtain the corresponding monthly population prediction set. Similarly, annual historical population characteristic data is combined with the preset medium-to-long-term prediction model set to obtain the corresponding annual population prediction set. Finally, the monthly population prediction set is processed to obtain the final monthly population prediction value, and the annual population prediction set is processed to obtain the final annual population prediction value. This achieves regional population dynamic monitoring and prediction technology based on spatiotemporal big data.

[0024] According to an embodiment of the present invention, the acquisition of historical population characteristic data of a preset area includes monthly historical population characteristic data and annual historical population characteristic data, including: Obtain historical population characteristic data for the preset area, including monthly and annual historical population characteristic data; The monthly historical demographic data includes monthly resident population statistics, monthly natural population growth rate, monthly mechanical population growth rate, and monthly registered regional resident population. The annual historical demographic data includes annual resident population statistics, annual natural population growth rate, annual mechanical growth rate, annual regional population density, annual regional GDP, and annual regional employment rate.

[0025] It is important to note that, in order to achieve regional population dynamic monitoring and prediction based on spatiotemporal big data, it is first necessary to acquire historical population characteristic data of the preset monitoring area to construct a complete historical data foundation. The data covers two time scales: monthly and annual, to ensure the comprehensiveness and timeliness of the data. Among them, the monthly historical population characteristic data focuses on short-term population dynamic changes, mainly including monthly resident population statistics, monthly natural population growth rate, monthly mechanical growth rate, and monthly regional permanent resident population registration numbers, which can accurately capture the monthly population fluctuation patterns and reflect the natural increase, decrease, and flow changes of the population in the short term. The annual historical population characteristic data focuses on long-term population development trends. In addition to annual resident population statistics, annual natural population growth rate, and annual mechanical growth rate, it also supplements annual regional population density, annual regional GDP, and annual regional employment rate. Combined with economic, employment, and other related factors, it comprehensively depicts the long-term distribution and development trend of the regional population.

[0026] According to an embodiment of the present invention, the step of obtaining a preset prediction model set, performing a filtering process to obtain a preset prediction model set that meets the prediction requirements, and marking them respectively as a preset short-term prediction model set and a preset medium- and long-term prediction model set, includes: Obtain the preset prediction model set, including the preset exponential smoothing model set, the preset ARIMA model set, the preset Prophet model set, the preset spatial lag model set, and the preset XGBoost model set; Obtain the average absolute percentage error corresponding to each preset prediction model in the preset prediction model set; Obtain the first preset average absolute percentage error threshold and the second preset average absolute percentage error threshold; The average absolute percentage error corresponding to the preset exponential smoothing model set, the preset ARIMA model set, and the preset Prophet model set is compared with the first preset average absolute percentage error threshold to obtain a first comparison result. Based on the first comparison results, a first set of preset prediction models that meet the prediction requirements is obtained and marked as a set of preset short-term prediction models; The average absolute percentage error corresponding to the preset ARIMA model set, preset Prophet model set, preset spatial lag model set and preset XGBoost model set are compared with the second preset average absolute percentage error threshold to obtain the second comparison result. Based on the second comparison results, a second preset prediction model set that meets the prediction requirements is obtained and marked as the preset medium- and long-term prediction model set.

[0027] It is important to note that, to achieve accuracy in short-term and medium-to-long-term regional population forecasts and adapt to forecasting needs at different time scales, a pre-defined forecasting model set must first be acquired and scientifically screened and categorized. The specific steps are as follows: First, acquire a pre-defined forecasting model set with multiple algorithm types. This model set comprehensively covers time series forecasting and spatiotemporal correlation forecasting scenarios, specifically including a pre-defined exponential smoothing model set, a pre-defined ARIMA model set, a pre-defined Prophet model set, a pre-defined spatial lag model set, and a pre-defined XGBoost model set. These models complement each other and adapt to different population data fluctuation characteristics, providing a sufficient model foundation for subsequent screening. Since forecasting accuracy is the core evaluation criterion for model screening, it is necessary to obtain the mean absolute percentage error (MAPE) for each pre-defined forecasting model in the pre-defined forecasting model set. This indicator accurately reflects the degree of forecasting deviation of each model and is the core basis for selecting models that meet the requirements. Simultaneously, considering the different accuracy requirements of short-term and medium-to-long-term population forecasts, two differentiated error thresholds are pre-defined: a first pre-defined mean absolute percentage error threshold and a second pre-defined mean absolute percentage error threshold. The first threshold adapts to the high-precision requirements of short-term forecasts, while the second... The threshold is adjusted to meet the reasonable requirements of medium- and long-term forecasting. Next, scenario-based screening is conducted based on the preset thresholds: For short-term population forecasting, the mean absolute percentage errors (MAEs) of the preset exponential smoothing model set, preset ARIMA model set, and preset Prophet model set are compared one by one with the first preset MAE threshold to obtain a first comparison result indicating whether each model meets the short-term forecasting accuracy requirements. Based on this result, models that meet the error standards and are suitable for short-term forecasting are selected to form the first preset forecasting model set, which is then marked as the preset short-term forecasting model set for subsequent monthly population forecasting. For medium- and long-term population forecasting, the MAEs of the preset ARIMA model set, preset Prophet model set, preset spatial lag model set, and preset XGBoost model set are compared one by one with the second preset MAE threshold to obtain a second comparison result indicating whether each model meets the medium- and long-term forecasting accuracy requirements. Based on this result, models that meet the error standards and are suitable for medium- and long-term forecasting are selected to form the second preset forecasting model set, which is then marked as the preset medium- and long-term forecasting model set for subsequent annual population forecasting.

[0028] According to an embodiment of the present invention, the step of collecting real-time population characteristic data of the preset area, and processing it in conjunction with the monthly historical population characteristic data and the annual historical population characteristic data through the preset short-term prediction model set and the preset medium- and long-term prediction model set respectively, to obtain the corresponding monthly population prediction value set and annual population prediction value set, includes: Collect real-time population characteristic data of the preset area, including real-time population statistics, real-time birth rate, real-time death rate, real-time population inflow and real-time population outflow; The real-time population characteristic data and the monthly historical population characteristic data are processed by the preset short-term prediction model set to obtain the corresponding monthly population prediction value set. The real-time population characteristic data and the annual historical population characteristic data are processed by the preset medium- and long-term prediction model set to obtain the corresponding annual population prediction value set.

[0029] It is important to note that after completing the historical demographic data processing and predictive model set selection and classification, real-time demographic data for the preset region must first be collected to provide immediate support for multi-scale prediction. The collection process follows the principles of accuracy and real-time monitoring, specifically including real-time population statistics, real-time birth rate, real-time death rate, real-time population inflow, and real-time population outflow, ensuring data integrity and consistency to comprehensively capture the real-time population changes in the region. Next, for short-term prediction, the real-time demographic data is correlated and calibrated with monthly historical demographic data before being input into the preset short-term prediction model set. Each model in the set independently processes the dataset: the preset exponential smoothing model, A... The RIMA and Prophet models, each based on their own algorithms and combined with data feature calculations, obtain corresponding monthly population forecasts. The forecasts output by all models are aggregated to form the corresponding monthly population forecast set. For medium- and long-term forecasts, real-time population feature data is deeply integrated with annual historical population feature data and then input into a preset medium- and long-term forecast model set. Similarly, each model in the set processes the data separately: the ARIMA model, Prophet model, spatial lag model, and XGBoost model leverage their algorithmic advantages to independently calculate the corresponding annual population forecasts. All forecasts are then integrated to form the corresponding annual population forecast set.

[0030] According to an embodiment of the present invention, the step of processing the monthly population forecast set to obtain the final monthly population forecast includes: Statistical processing is performed on the monthly population forecast set to obtain the mean and standard deviation of the population forecasts. A preset reasonable prediction threshold range is obtained based on the mean and standard deviation of the predicted population values. The third comparison result is obtained by comparing each monthly population forecast value in the monthly population forecast set with the preset reasonable forecast threshold range. Based on the third comparison result, a set of reasonable values ​​for monthly population forecasts is obtained; The final monthly population forecast value is obtained by weighting the set of reasonable monthly population forecast values.

[0031] It is important to note that after obtaining the monthly population forecast set, multiple steps are required to filter reasonable values ​​and calculate the final forecast result. The specific steps are as follows: First, systematically statistically process the monthly population forecast set to accurately calculate the mean population forecast value of all forecast values ​​within the set, and simultaneously calculate the standard deviation of the population forecast values ​​to reflect the central tendency and dispersion of all model forecast results. Based on the calculated mean and standard deviation of the population forecast values, a preset reasonable forecast threshold range is set and obtained. For example, the preset reasonable forecast threshold range is set as [mean population forecast value - 3 * standard deviation of population forecast value, population forecast value]... The range [mean + 3 * standard deviation of population forecast] effectively eliminates outlier forecasts, ensuring the reliability of subsequent results. Then, each forecast value in the monthly population forecast set is compared one by one with a preset reasonable forecast threshold range to obtain a third comparison result indicating whether each forecast value falls within a reasonable range. Based on this third comparison result, all forecast values ​​within the reasonable threshold range are selected to form a reasonable monthly population forecast value set. Finally, combining the prediction accuracy weights of each short-term forecast model, a weighted calculation is performed on each reasonable forecast value in the reasonable monthly population forecast value set to obtain an accurate and reliable final monthly population forecast value.

[0032] According to an embodiment of the present invention, the step of processing the annual population forecast set to obtain the final value of the annual population forecast includes: The preset residual threshold range is obtained based on the preset long-term trend residual method; The fourth comparison result is obtained by comparing the annual population forecast values ​​of each year in the set of annual population forecast values ​​with the preset residual threshold interval. Based on the fourth comparison result, a set of reasonable values ​​for annual population forecasts is obtained; The final annual population forecast is obtained by weighting the set of reasonable annual population forecast values.

[0033] It is important to note that, to ensure the accuracy and reliability of the annual population forecast results, the previously obtained annual population forecast set needs to undergo multi-step screening and integration processing. The specific steps are as follows: First, using the preset long-term trend residual method, combined with the long-term development pattern of regional population and the error characteristics of historical data, the preset residual threshold range is accurately obtained through residual analysis, threshold calibration, and other operations. This range is used to effectively identify abnormal data in the annual population forecast set and avoid the impact of extreme forecast results on the final value. Next, the annual forecast value output by each model is compared with the preset residual threshold range one by one to determine whether each forecast value is within a reasonable residual range, thereby obtaining the comparison results corresponding to each forecast value. After summarizing, a fourth comparison result is formed. Based on the fourth comparison result, all annual population forecast values ​​within the preset residual threshold range are selected, and abnormally deviated forecast data are removed to form a reasonable set of annual population forecast values. Finally, combining the prediction accuracy weights of each model in the preset medium- and long-term prediction model set, each reasonable forecast value in the reasonable set of annual population forecast values ​​is weighted and calculated to finally obtain an accurate and reliable final value of the annual population forecast.

[0034] Please refer to Figure 2 , Figure 2 This is a schematic diagram illustrating a regional population dynamics monitoring and prediction method based on spatiotemporal big data provided in this application embodiment. According to this embodiment, for example, a systematic statistical processing is performed on the monthly population prediction set to calculate the mean μ and standard deviation σ of all predicted values ​​within the set. Based on the calculated mean and standard deviation, a preset reasonable prediction threshold range is set and obtained. This preset reasonable prediction threshold range is set to [μ-3σ, μ+3σ], which effectively eliminates abnormal prediction values, ensuring the reliability of subsequent results. Subsequently, each predicted value in the monthly population prediction set is compared one by one with the preset reasonable prediction threshold range to obtain a third comparison result indicating whether each predicted value falls within the reasonable range. Based on the third comparison result, all predicted values ​​within the reasonable threshold range are selected to form a reasonable monthly population prediction value set. Finally, combining the prediction accuracy weights of each short-term prediction model, a weighted calculation is performed on each reasonable prediction value in the reasonable monthly population prediction value set to obtain an accurate and reliable final monthly population prediction value.

[0035] Secondly, the present invention also discloses a regional population dynamics monitoring and prediction system based on spatiotemporal big data, including a memory and a processor. The memory includes a method program for regional population dynamics monitoring and prediction based on spatiotemporal big data. When the method program for regional population dynamics monitoring and prediction based on spatiotemporal big data is executed by the processor, it performs the following steps: Obtain historical population characteristic data for the preset area, including monthly and annual historical population characteristic data; Obtain a preset prediction model set, perform filtering to obtain a preset prediction model set that meets the prediction requirements, and mark them as the preset short-term prediction model set and the preset medium- and long-term prediction model set respectively. Real-time population characteristic data of the preset area is collected, and combined with the monthly historical population characteristic data and the annual historical population characteristic data, it is processed by the preset short-term prediction model set and the preset medium- and long-term prediction model set respectively to obtain the corresponding monthly population prediction value set and annual population prediction value set. The final monthly population forecast value is obtained by processing the set of monthly population forecast values. The final annual population forecast is obtained by processing the set of annual population forecast values.

[0036] It is important to note that traditional population monitoring relies on censuses and sampling surveys, which suffer from poor timeliness, low spatial resolution, and high labor costs. Existing forecasting technologies often employ single models or single-scale data, failing to differentiate between short-term and medium-to-long-term forecasting needs, and lacking deep integration of historical data with real-time spatiotemporal big data. This leads to forecasting biases and an inability to balance monthly and annual forecast accuracy. To address these issues, the first step is to acquire historical population characteristic data for the predefined region, including monthly and annual historical population characteristic data. Monthly historical population characteristic data includes monthly resident population statistics, monthly natural population growth rate, monthly mechanical population growth rate, and monthly regional permanent resident registration numbers. Annual historical population characteristic data includes annual resident population statistics, annual natural population growth rate, annual mechanical population growth rate, annual regional population density, annual regional GDP, and annual regional employment data. The system acquires a set of preset prediction models, including preset exponential smoothing models, preset ARIMA models, preset Prophet models, preset spatial lag models, and preset XGBoost models. These models are then filtered to obtain the preset prediction models that meet the prediction requirements, and are labeled as preset short-term prediction models and preset medium-to-long-term prediction models, respectively. Next, real-time population characteristic data for a preset region is collected and combined with monthly historical population characteristic data, processed using the preset short-term prediction model set to obtain the corresponding monthly population prediction set. Similarly, annual historical population characteristic data is combined with the preset medium-to-long-term prediction model set to obtain the corresponding annual population prediction set. Finally, the monthly population prediction set is processed to obtain the final monthly population prediction value, and the annual population prediction set is processed to obtain the final annual population prediction value. This achieves regional population dynamic monitoring and prediction technology based on spatiotemporal big data.

[0037] According to an embodiment of the present invention, the acquisition of historical population characteristic data of a preset area includes monthly historical population characteristic data and annual historical population characteristic data, including: Obtain historical population characteristic data for the preset area, including monthly and annual historical population characteristic data; The monthly historical demographic data includes monthly resident population statistics, monthly natural population growth rate, monthly mechanical population growth rate, and monthly registered regional resident population. The annual historical demographic data includes annual resident population statistics, annual natural population growth rate, annual mechanical growth rate, annual regional population density, annual regional GDP, and annual regional employment rate.

[0038] It is important to note that, in order to achieve regional population dynamic monitoring and prediction based on spatiotemporal big data, it is first necessary to acquire historical population characteristic data of the preset monitoring area to construct a complete historical data foundation. The data covers two time scales: monthly and annual, to ensure the comprehensiveness and timeliness of the data. Among them, the monthly historical population characteristic data focuses on short-term population dynamic changes, mainly including monthly resident population statistics, monthly natural population growth rate, monthly mechanical growth rate, and monthly regional permanent resident population registration numbers, which can accurately capture the monthly population fluctuation patterns and reflect the natural increase, decrease, and flow changes of the population in the short term. The annual historical population characteristic data focuses on long-term population development trends. In addition to annual resident population statistics, annual natural population growth rate, and annual mechanical growth rate, it also supplements annual regional population density, annual regional GDP, and annual regional employment rate. Combined with economic, employment, and other related factors, it comprehensively depicts the long-term distribution and development trend of the regional population.

[0039] According to an embodiment of the present invention, the step of obtaining a preset prediction model set, performing a filtering process to obtain a preset prediction model set that meets the prediction requirements, and marking them respectively as a preset short-term prediction model set and a preset medium- and long-term prediction model set, includes: Obtain the preset prediction model set, including the preset exponential smoothing model set, the preset ARIMA model set, the preset Prophet model set, the preset spatial lag model set, and the preset XGBoost model set; Obtain the average absolute percentage error corresponding to each preset prediction model in the preset prediction model set; Obtain the first preset average absolute percentage error threshold and the second preset average absolute percentage error threshold; The average absolute percentage error corresponding to the preset exponential smoothing model set, the preset ARIMA model set, and the preset Prophet model set is compared with the first preset average absolute percentage error threshold to obtain a first comparison result. Based on the first comparison results, a first set of preset prediction models that meet the prediction requirements is obtained and marked as a set of preset short-term prediction models; The average absolute percentage error corresponding to the preset ARIMA model set, preset Prophet model set, preset spatial lag model set and preset XGBoost model set are compared with the second preset average absolute percentage error threshold to obtain the second comparison result. Based on the second comparison results, a second preset prediction model set that meets the prediction requirements is obtained and marked as the preset medium- and long-term prediction model set.

[0040] It is important to note that, to achieve accuracy in short-term and medium-to-long-term regional population forecasts and adapt to forecasting needs at different time scales, a pre-defined forecasting model set must first be acquired and scientifically screened and categorized. The specific steps are as follows: First, acquire a pre-defined forecasting model set with multiple algorithm types. This model set comprehensively covers time series forecasting and spatiotemporal correlation forecasting scenarios, specifically including a pre-defined exponential smoothing model set, a pre-defined ARIMA model set, a pre-defined Prophet model set, a pre-defined spatial lag model set, and a pre-defined XGBoost model set. These models complement each other and adapt to different population data fluctuation characteristics, providing a sufficient model foundation for subsequent screening. Since forecasting accuracy is the core evaluation criterion for model screening, it is necessary to obtain the mean absolute percentage error (MAPE) for each pre-defined forecasting model in the pre-defined forecasting model set. This indicator accurately reflects the degree of forecasting deviation of each model and is the core basis for selecting models that meet the requirements. Simultaneously, considering the different accuracy requirements of short-term and medium-to-long-term population forecasts, two differentiated error thresholds are pre-defined: a first pre-defined mean absolute percentage error threshold and a second pre-defined mean absolute percentage error threshold. The first threshold adapts to the high-precision requirements of short-term forecasts, while the second... The threshold is adjusted to meet the reasonable requirements of medium- and long-term forecasting. Next, scenario-based screening is conducted based on the preset thresholds: For short-term population forecasting, the mean absolute percentage errors (MAEs) of the preset exponential smoothing model set, preset ARIMA model set, and preset Prophet model set are compared one by one with the first preset MAE threshold to obtain a first comparison result indicating whether each model meets the short-term forecasting accuracy requirements. Based on this result, models that meet the error standards and are suitable for short-term forecasting are selected to form the first preset forecasting model set, which is then marked as the preset short-term forecasting model set for subsequent monthly population forecasting. For medium- and long-term population forecasting, the MAEs of the preset ARIMA model set, preset Prophet model set, preset spatial lag model set, and preset XGBoost model set are compared one by one with the second preset MAE threshold to obtain a second comparison result indicating whether each model meets the medium- and long-term forecasting accuracy requirements. Based on this result, models that meet the error standards and are suitable for medium- and long-term forecasting are selected to form the second preset forecasting model set, which is then marked as the preset medium- and long-term forecasting model set for subsequent annual population forecasting.

[0041] According to an embodiment of the present invention, the step of collecting real-time population characteristic data of the preset area, and processing it in conjunction with the monthly historical population characteristic data and the annual historical population characteristic data through the preset short-term prediction model set and the preset medium- and long-term prediction model set respectively, to obtain the corresponding monthly population prediction value set and annual population prediction value set, includes: Collect real-time population characteristic data of the preset area, including real-time population statistics, real-time birth rate, real-time death rate, real-time population inflow and real-time population outflow; The real-time population characteristic data and the monthly historical population characteristic data are processed by the preset short-term prediction model set to obtain the corresponding monthly population prediction value set. The real-time population characteristic data and the annual historical population characteristic data are processed by the preset medium- and long-term prediction model set to obtain the corresponding annual population prediction value set.

[0042] It is important to note that after completing the historical demographic data processing and predictive model set selection and classification, real-time demographic data for the preset region must first be collected to provide immediate support for multi-scale prediction. The collection process follows the principles of accuracy and real-time monitoring, specifically including real-time population statistics, real-time birth rate, real-time death rate, real-time population inflow, and real-time population outflow, ensuring data integrity and consistency to comprehensively capture the real-time population changes in the region. Next, for short-term prediction, the real-time demographic data is correlated and calibrated with monthly historical demographic data before being input into the preset short-term prediction model set. Each model in the set independently processes the dataset: the preset exponential smoothing model, A... The RIMA and Prophet models, each based on their own algorithms and combined with data feature calculations, obtain corresponding monthly population forecasts. The forecasts output by all models are aggregated to form the corresponding monthly population forecast set. For medium- and long-term forecasts, real-time population feature data is deeply integrated with annual historical population feature data and then input into a preset medium- and long-term forecast model set. Similarly, each model in the set processes the data separately: the ARIMA model, Prophet model, spatial lag model, and XGBoost model leverage their algorithmic advantages to independently calculate the corresponding annual population forecasts. All forecasts are then integrated to form the corresponding annual population forecast set.

[0043] According to an embodiment of the present invention, the step of processing the monthly population forecast set to obtain the final monthly population forecast includes: Statistical processing is performed on the monthly population forecast set to obtain the mean and standard deviation of the population forecasts. A preset reasonable prediction threshold range is obtained based on the mean and standard deviation of the predicted population values. The third comparison result is obtained by comparing each monthly population forecast value in the monthly population forecast set with the preset reasonable forecast threshold range. Based on the third comparison result, a set of reasonable values ​​for monthly population forecasts is obtained; The final monthly population forecast value is obtained by weighting the set of reasonable monthly population forecast values.

[0044] It is important to note that after obtaining the monthly population forecast set, multiple steps are required to filter reasonable values ​​and calculate the final forecast result. The specific steps are as follows: First, systematically statistically process the monthly population forecast set to accurately calculate the mean population forecast of all forecasts within the set, and simultaneously calculate the standard deviation of the population forecast to reflect the central tendency and dispersion of all model forecast results; based on the calculated mean and standard deviation of the population forecast, a preset reasonable forecast threshold range is set and obtained, for example, the preset reasonable forecast threshold range is set as [mean population forecast]... Standard deviation of population projections, mean of population projections The standard deviation of the population forecast is used to effectively eliminate outlier forecasts and ensure the reliability of subsequent results. Then, each forecast value in the monthly population forecast set is compared with a preset reasonable forecast threshold range to obtain a third comparison result indicating whether each forecast value falls within a reasonable range. Based on this third comparison result, all forecast values ​​within the reasonable threshold range are selected to form a reasonable monthly population forecast set. Finally, combining the prediction accuracy weights of each short-term forecast model, a weighted calculation is performed on each reasonable forecast value in the reasonable monthly population forecast set to obtain an accurate and reliable final monthly population forecast value.

[0045] According to an embodiment of the present invention, the step of processing the annual population forecast set to obtain the final value of the annual population forecast includes: The preset residual threshold range is obtained based on the preset long-term trend residual method; The fourth comparison result is obtained by comparing the annual population forecast values ​​of each year in the set of annual population forecast values ​​with the preset residual threshold interval. Based on the fourth comparison result, a set of reasonable values ​​for annual population forecasts is obtained; The final annual population forecast is obtained by weighting the set of reasonable annual population forecast values.

[0046] It is important to note that, to ensure the accuracy and reliability of the annual population forecast results, the previously obtained annual population forecast set needs to undergo multi-step screening and integration processing. The specific steps are as follows: First, using the preset long-term trend residual method, combined with the long-term development pattern of regional population and the error characteristics of historical data, the preset residual threshold range is accurately obtained through residual analysis, threshold calibration, and other operations. This range is used to effectively identify abnormal data in the annual population forecast set and avoid the impact of extreme forecast results on the final value. Next, the annual forecast value output by each model is compared with the preset residual threshold range one by one to determine whether each forecast value is within a reasonable residual range, thereby obtaining the comparison results corresponding to each forecast value. After summarizing, a fourth comparison result is formed. Based on the fourth comparison result, all annual population forecast values ​​within the preset residual threshold range are selected, and abnormally deviated forecast data are removed to form a reasonable set of annual population forecast values. Finally, combining the prediction accuracy weights of each model in the preset medium- and long-term prediction model set, each reasonable forecast value in the reasonable set of annual population forecast values ​​is weighted and calculated to finally obtain an accurate and reliable final value of the annual population forecast.

[0047] This invention discloses a method and system for regional population dynamic monitoring and prediction based on spatiotemporal big data. It acquires historical population characteristic data of a preset region, including monthly and annual historical population characteristic data; acquires a preset prediction model set, performs filtering processing to obtain preset prediction model sets that meet prediction requirements, and labels them as preset short-term prediction model sets and preset medium-to-long-term prediction model sets, respectively; collects real-time population characteristic data of the preset region; and processes the monthly and annual historical population characteristic data using the preset short-term and medium-to-long-term prediction model sets, respectively, to obtain corresponding monthly and annual population prediction value sets. The final monthly population prediction value is obtained by processing the monthly population prediction value set, and the final annual population prediction value is obtained by processing the annual population prediction value set, thereby realizing the technology for regional population dynamic monitoring and prediction based on spatiotemporal big data.

[0048] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0049] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0050] In addition, in the various embodiments of the present invention, each functional unit can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0051] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0052] Alternatively, if the integrated units of this invention are implemented as software functional modules and sold or used as independent products, they can also be stored in a readable storage medium. Based on this understanding, the technical solutions of the embodiments of this invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROM, RAM, magnetic disks, or optical disks.

Claims

1. A method for regional population dynamics monitoring and prediction based on spatiotemporal big data, characterized in that, Includes the following steps: Obtain historical population characteristic data for the preset area, including monthly and annual historical population characteristic data; Obtain a preset prediction model set, perform filtering to obtain a preset prediction model set that meets the prediction requirements, and mark them as the preset short-term prediction model set and the preset medium- and long-term prediction model set respectively. Real-time population characteristic data of the preset area is collected, and combined with the monthly historical population characteristic data and the annual historical population characteristic data, it is processed by the preset short-term prediction model set and the preset medium- and long-term prediction model set respectively to obtain the corresponding monthly population prediction value set and annual population prediction value set. The final monthly population forecast value is obtained by processing the set of monthly population forecast values. The final annual population forecast is obtained by processing the set of annual population forecast values.

2. The method for regional population dynamic monitoring and prediction based on spatiotemporal big data according to claim 1, characterized in that, The acquisition of historical population characteristic data for a preset area includes monthly historical population characteristic data and annual historical population characteristic data, including: Obtain historical population characteristic data for the preset area, including monthly and annual historical population characteristic data; The monthly historical demographic data includes monthly resident population statistics, monthly natural population growth rate, monthly mechanical population growth rate, and monthly registered regional resident population. The annual historical demographic data includes annual resident population statistics, annual natural population growth rate, annual mechanical growth rate, annual regional population density, annual regional GDP, and annual regional employment rate.

3. The method for regional population dynamic monitoring and prediction based on spatiotemporal big data according to claim 1, characterized in that, The process of acquiring a preset prediction model set, performing a filtering process to obtain a preset prediction model set that meets the prediction requirements, and marking them as a preset short-term prediction model set and a preset medium- and long-term prediction model set, includes: Obtain the preset prediction model set, including the preset exponential smoothing model set, the preset ARIMA model set, the preset Prophet model set, the preset spatial lag model set, and the preset XGBoost model set; Obtain the average absolute percentage error corresponding to each preset prediction model in the preset prediction model set; Obtain the first preset average absolute percentage error threshold and the second preset average absolute percentage error threshold; The average absolute percentage error corresponding to the preset exponential smoothing model set, the preset ARIMA model set, and the preset Prophet model set is compared with the first preset average absolute percentage error threshold to obtain a first comparison result. Based on the first comparison results, a first set of preset prediction models that meet the prediction requirements is obtained and marked as a set of preset short-term prediction models; The average absolute percentage error corresponding to the preset ARIMA model set, preset Prophet model set, preset spatial lag model set and preset XGBoost model set are compared with the second preset average absolute percentage error threshold to obtain the second comparison result. Based on the second comparison results, a second preset prediction model set that meets the prediction requirements is obtained and marked as the preset medium- and long-term prediction model set.

4. The method for regional population dynamic monitoring and prediction based on spatiotemporal big data according to claim 1, characterized in that, The process involves collecting real-time population characteristic data from the preset area, combining it with monthly and annual historical population characteristic data, and then processing it using the preset short-term prediction model set and preset medium-to-long-term prediction model set to obtain corresponding monthly and annual population prediction value sets, including: Collect real-time population characteristic data of the preset area, including real-time population statistics, real-time birth rate, real-time death rate, real-time population inflow and real-time population outflow; The real-time population characteristic data and the monthly historical population characteristic data are processed by the preset short-term prediction model set to obtain the corresponding monthly population prediction value set. The real-time population characteristic data and the annual historical population characteristic data are processed by the preset medium- and long-term prediction model set to obtain the corresponding annual population prediction value set.

5. The method for regional population dynamic monitoring and prediction based on spatiotemporal big data according to claim 1, characterized in that, The step of processing the monthly population forecast set to obtain the final monthly population forecast includes: Statistical processing is performed on the monthly population forecast set to obtain the mean and standard deviation of the population forecasts. A preset reasonable prediction threshold range is obtained based on the mean and standard deviation of the predicted population values. The third comparison result is obtained by comparing each monthly population forecast value in the monthly population forecast set with the preset reasonable forecast threshold range. Based on the third comparison result, a set of reasonable values ​​for monthly population forecasts is obtained; The final monthly population forecast value is obtained by weighting the set of reasonable monthly population forecast values.

6. The method for regional population dynamic monitoring and prediction based on spatiotemporal big data according to claim 1, characterized in that, The process of obtaining the final annual population forecast value based on the set of annual population forecast values ​​includes: The preset residual threshold range is obtained based on the preset long-term trend residual method; The fourth comparison result is obtained by comparing the annual population forecast values ​​of each year in the set of annual population forecast values ​​with the preset residual threshold interval. Based on the fourth comparison result, a set of reasonable values ​​for annual population forecasts is obtained; The final annual population forecast is obtained by weighting the set of reasonable annual population forecast values.

7. A regional population dynamics monitoring and prediction system based on spatiotemporal big data, characterized in that, The system includes a memory and a processor. The memory contains a program for a regional population dynamics monitoring and prediction method based on spatiotemporal big data. When the processor executes the program for the regional population dynamics monitoring and prediction method based on spatiotemporal big data, it performs the following steps: Obtain historical population characteristic data for the preset area, including monthly and annual historical population characteristic data; Obtain a preset prediction model set, perform filtering to obtain a preset prediction model set that meets the prediction requirements, and mark them as the preset short-term prediction model set and the preset medium- and long-term prediction model set respectively. Real-time population characteristic data of the preset area is collected, and combined with the monthly historical population characteristic data and the annual historical population characteristic data, it is processed by the preset short-term prediction model set and the preset medium- and long-term prediction model set respectively to obtain the corresponding monthly population prediction value set and annual population prediction value set. The final monthly population forecast value is obtained by processing the set of monthly population forecast values. The final annual population forecast is obtained by processing the set of annual population forecast values.

8. The regional population dynamics monitoring and prediction system based on spatiotemporal big data according to claim 7, characterized in that, The acquisition of historical population characteristic data for a preset area includes monthly historical population characteristic data and annual historical population characteristic data, including: Obtain historical population characteristic data for the preset area, including monthly and annual historical population characteristic data; The monthly historical demographic data includes monthly resident population statistics, monthly natural population growth rate, monthly mechanical population growth rate, and monthly registered regional resident population. The annual historical demographic data includes annual resident population statistics, annual natural population growth rate, annual mechanical growth rate, annual regional population density, annual regional GDP, and annual regional employment rate.

9. The regional population dynamics monitoring and prediction system based on spatiotemporal big data according to claim 7, characterized in that, The process of acquiring a preset prediction model set, performing a filtering process to obtain a preset prediction model set that meets the prediction requirements, and marking them as a preset short-term prediction model set and a preset medium- and long-term prediction model set, includes: Obtain the preset prediction model set, including the preset exponential smoothing model set, the preset ARIMA model set, the preset Prophet model set, the preset spatial lag model set, and the preset XGBoost model set; Obtain the average absolute percentage error corresponding to each preset prediction model in the preset prediction model set; Obtain the first preset average absolute percentage error threshold and the second preset average absolute percentage error threshold; The average absolute percentage error corresponding to the preset exponential smoothing model set, the preset ARIMA model set, and the preset Prophet model set is compared with the first preset average absolute percentage error threshold to obtain a first comparison result. Based on the first comparison results, a first set of preset prediction models that meet the prediction requirements is obtained and marked as a set of preset short-term prediction models; The average absolute percentage error corresponding to the preset ARIMA model set, preset Prophet model set, preset spatial lag model set and preset XGBoost model set are compared with the second preset average absolute percentage error threshold to obtain the second comparison result. Based on the second comparison results, a second preset prediction model set that meets the prediction requirements is obtained and marked as the preset medium- and long-term prediction model set.

10. The regional population dynamics monitoring and prediction system based on spatiotemporal big data according to claim 7, characterized in that, The process involves collecting real-time population characteristic data from the preset area, combining it with monthly and annual historical population characteristic data, and then processing it using the preset short-term prediction model set and preset medium-to-long-term prediction model set to obtain corresponding monthly and annual population prediction value sets, including: Collect real-time population characteristic data of the preset area, including real-time population statistics, real-time birth rate, real-time death rate, real-time population inflow and real-time population outflow; The real-time population characteristic data and the monthly historical population characteristic data are processed by the preset short-term prediction model set to obtain the corresponding monthly population prediction value set. The real-time population characteristic data and the annual historical population characteristic data are processed by the preset medium- and long-term prediction model set to obtain the corresponding annual population prediction value set.

Images