Identity association method and device based on track similarity, equipment and storage medium

Abstract

The invention discloses an identity association method and device based on track similarity, equipment and a storage medium, and the method comprises the steps: carrying out the serialization of to-be-analyzed track data of each system, and obtaining each track sequence in each system; calculating detail similarity and overall similarity between each track sequence in each system and each track sequence in other systems; for each track sequence in each system, performing weighted calculation according to the detail similarity and the overall similarity between the track sequence in each systemand each track sequence in other systems to obtain weighted track similarity between the track sequence in each system and each track sequence in other systems; and according to the weighted trajectory similarity between each trajectory sequence in each system and each trajectory sequence in other systems, associating the user identity information of the mutually similar trajectory sequences between different systems. By adopting the method and the device, the identity information corresponding to the mutually similar tracks among different systems can be accurately associated.

Description

technical field [0001] The present invention relates to the technical field of data mining, in particular to an identity association method, device, equipment and storage medium based on trajectory similarity. Background technique [0002] With the development of technology, information about people's activities is now recorded by various systems, such as user trajectories recorded by mobile phone operators, user movement trajectories recorded by mobile phone APPs, vehicle movement trajectories recorded by car navigation, and bank records. The bank card swiping information below, the hotel records the accommodation information of the residents, and so on. The identity IDs recorded in these information are different. For example, the identity ID recorded by the mobile phone operator is the mobile phone number, the mobile APP records the APP account opening ID, the identity ID recorded by the car navigation is the machine code of the navigator, and the identity ID recorded by ...

AI Technical Summary

Problems solved by technology

[0004] However, the inventors found in the process of implementing the present invention that because the DTW algorithm itself is more suitable for the characteristics of the overall trend comparison between trajectories, the existing identity association methods are only based on the overall characteristics between the target trajectory and all trajectories to be compared The degree of similarity between trajectories is used for identity association, while the similarity of detailed features between trajectories is ignored, which affects the accuracy of identity association.

Images