Big data-based identification method and system for objects with similar traces

Abstract

The invention belongs to the field of data analysis, and relates to a big data-based identification method and system for objects with similar traces. A spatial normalization module, a data arrangement module, a spatial point position sequence generation module, a behavior trace sequence extraction module, a behavior matrix generation module and a behavior matrix calculation module are connected in sequence, and a trace identifier dictionary is constructed for original trace position data; the data is introduced in a standardized manner according to the trace identifier dictionary; trace pointposition data belonging to objects is integrated; trace point position sequences of the objects in a time window are constructed; displacement trace sequences are generated for the trace point position sequences of the objects; a behavior calculation matrix is constructed for calculating trace similarity; and results with behavior similarity meeting the condition are output. By obtaining and analyzing spatial position moving information of different objects and calculating the similarity between spatial position changes of the different objects by utilizing an algorithm, a similarity relationship between the objects is explored, and the objects and groups with similar behaviors are identified and searched for.

Description

technical field [0001] The invention belongs to the technical field of data analysis, in particular to the technical field of behavioral similarity data analysis, and in particular to a method and system for identifying similar objects based on big data. Background technique [0002] In the fields of e-commerce and public security, it is hoped that by analyzing the data collected by a large number of data collection devices scattered around or the network nodes visited by users, it is hoped to find objects with similar behaviors to the target person or group. When using the traditional precise location-based track matching method for identification, there are often problems such as time-consuming, too much suspected data, and difficult to filter, especially in today's situation of geometric growth of data volume, the traditional method is based on the limitations of localized centralized calculation Increasingly prominent, it is becoming more and more difficult to meet the n...

AI Technical Summary

Problems solved by technology

When using the traditional precise location-based track matching method for identification, there are often problems such as time-consuming, too much suspected data, and difficult to filter, especially in today's situation of geometric growth of data volume, the traditional method is based on the limitations of localized centralized calculation Increasingly prominent, increasingly difficult to meet needs

Images