Method, device, equipment and medium for predicting personnel flow based on signaling data

Abstract

The invention discloses a method, device, equipment and medium for predicting personnel flow based on signaling data. The method includes: acquiring long-sequence signaling data and geographic information data at the same period in history; Divide multiple time periods, establish and solve flow forecast models for different periods, and obtain the initial flow forecast results without special event influence; when there is special event influence, calculate the origin attenuation coefficient and destination attenuation coefficient, and based on the coefficient Correct the initial traffic forecast results to obtain the first traffic forecast results under the influence of special events; dynamically adjust the origin attenuation coefficient and destination attenuation coefficient with the data update, and adjust the first traffic forecast results according to the coefficients to obtain special events Influenced by the second traffic forecast results. This method can use mobile phone signaling data with a large sample size and wide coverage to predict the number and spatio-temporal characteristics of migrant workers during major holidays.

Description

technical field [0001] The present invention relates to the field of big data technology, in particular to a method, device, equipment and medium for predicting personnel flow based on signaling data. Background technique [0002] The large-scale flow of migrant workers is of great significance to the prevention and control of infectious diseases, transportation management, labor security, housing and public service allocation, economic consumption vitality, social security governance, urbanization pattern, resource consumption, and environmental protection. , is the focus of policy formulation on population flow and migration, regional and urban governance, and spatial planning around the world. Therefore, how to use big data to simulate and predict the flow of migrant workers with high precision is a prerequisite for formulating relevant policies scientifically and rationally. [0003] In the existing technology, there is a personnel flow analysis method based on geograph...

AI Technical Summary

Problems solved by technology

[0003] In the existing technology, there is a personnel flow analysis method based on geographic location service technology. First, this technology only uses big data to analyze the current situation of personnel flow. The application of big data mainly stays at the display level and fails to extract the information of personnel flow. The general law of the spatio-temporal change process does not realize the function of predicting the future flow of people; second, geographic location services require end users to actively initiate service requests before they can obtain relevant spatio-temporal data. There are a large number of platforms that provide this service, so the degree of data coverage It needs to be improved, and a large number of spatial flow behaviors have not been effectively recorded; third, the data obtained by this technology does not have user ID information, so it cannot identify the user's complete travel, but only counts the flow of people between cities

[0004] There is also a major holiday OD traffic forecast method based on linear and nonlinear regression. Although this method is simple, effective and reasonable, the determination of explanatory variables in the multivariate model depends on empirical judgment and subjective selection, and the selection of explanatory variables will be difficult. have some influence on the predictive effect

In addition, the flow data of expressway exits used in this method can only reflect the OD flow of a single traffic mode on the expressway during major holidays, but cannot reflect the OD volume of personnel flow during holidays that integrate multiple modes of traffic.

Moreover, this prediction method is mostly applicable to a small prediction area. For a larger range, such as the traffic OD volume prediction of the whole country, the calculation process of this method will be more complicated.

Images