Consignment smoke-related prediction method fusing space-time and network topology characteristics, medium and equipment

Abstract

The invention belongs to a smoke-related prediction method, and provides a consignment smoke-related prediction method, a computer readable storage medium and equipment in order to solve the technical problems that an existing crime prediction method is not suitable for consignment smoke-related big data analysis and has defects in the aspects of feature calculation and big data feature analysis. Big data and an artificial intelligence technology are used in posting tobacco crime analysis, a new concept of a posting spatio-temporal mode is provided, spatio-temporal characteristics of posting big data are calculated according to the new concept, a posting network is described by using a weighted directed graph, network topology characteristics of the posting network are calculated, a posting relationship is described according to the network topology characteristics, a feature selection method is used, and the posting tobacco crime analysis is realized. According to the method, the optimal characteristics are selected, and a posting smoke-related prediction model is constructed by combining optimal characteristic calculation and a classifier method, so that a result is obtained in the aspect of studying and judging the precision of unknown posting data, and the feasibility and effectiveness of the prediction method are proved.

Description

technical field [0001] The invention belongs to a smoke-related prediction method, and in particular relates to a delivery-related smoke-related prediction method, medium and equipment integrating space-time and network topology features. Background technique [0002] In recent years, with the rapid development of the delivery industry and the e-commerce industry, relying on the mobile social platform, the delivery package is used to disguise the transportation of counterfeit cigarettes and private cigarettes, and the electronic settlement platform is used for real-time settlement. Tobacco crime has become an important criminal model that damages the development of the tobacco industry. [0003] In his dissertation "Urban Theft Crime Prediction Method Fusion of LSTM and ST-GCN", Han Xinge from the People's Public Security University of China analyzed and extracted the time-series and spatiotemporal characteristics of the number of urban theft crimes, combined with external e...

AI Technical Summary

Problems solved by technology

[0007] In order to solve the technical problem that the existing crime prediction method is not suitable for the big data analysis of mailing and delivery tobacco, and there are insufficient features in feature calculation and big data feature analysis, the present invention provides a delivery forecasting method that integrates space-time and network topology features. Smoke prediction method, medium and equipment

Images