A social relationship judgment method based on moving trajectory

Abstract

The invention relates to a method for judging social relations based on moving tracks. The method firstly performs data cleaning on the collected positioning data of residents to obtain track point information of moving objects; then uses the improved density-based clustering algorithm DBSCAN to Extract the stay points in the trajectory, and use the POI database to semanticize the stay points to obtain the semantic trajectory of the moving object; finally, analyze the co-occurrence of the semantic trajectories of different moving objects to obtain the co-occurrence characteristics between different moving objects, The classification algorithm SVM in machine learning is used to complete the social relationship judgment of different moving objects. Based on the positioning data, the present invention judges the social relationship between mobile users from the perspective of physical interaction of mobile users in the real world, can accurately complete group relationship division, and provides an effective solution for accurate description of user portraits and group monitoring and tracking in the public security field. data reference.

Description

technical field [0001] The invention relates to the field of research and application of spatio-temporal data mining, in particular to a method for judging social relations based on moving tracks. Background technique [0002] In recent years, with the continuous development of GPS, satellite communication and other technologies and the widespread use of mobile smart devices with positioning functions, the tracking accuracy and precision of moving objects have been significantly improved. The rapid development of technologies such as mobile Internet, big data, cloud computing, sensor networks, and computer storage enables these trajectory data to be stored persistently, forming spatiotemporal trajectory data. These data not only record the location information of people in time series, but also imply the behavior characteristics of individuals themselves, as well as the relationship between people and people, people and cities, and people and society. In the research on the...

AI Technical Summary

Problems solved by technology

[0003] Existing trajectory data analysis is mainly focused on the level of motion law mining, without extracting deep knowledge about users hidden in trajectory data, such as social relations, resulting in low abstraction of mining results and difficulty in expressing high-level semantics of users

Social relationship judgment is one of the research hotspots in social network analysis. The existing mobile social network still provides traditional online social network services based on users’ online virtual social relationships, such as using mobile phone data and network social data for calculation. There is no research yet. Calculated from the actual geographic track perspective

Images