Evolution Calculation Of The Moving Object Social Relationship In Location-based Social Network

The advancement of location acquisition technology and the popularity of mobile devices have played an important role in the development of location-based social networks.In location-based social networks,people can upload geo-tagged video,photos or text,share their current location information,or share their travel routes by Posting GPS track records.While using various mobile devices,users generate a large amount of data related to their own location information.Based on these data,the mapping relationship between users’ online social contact and offline activities can be built,which can promote the development of location services.Based on the data of location-based social networks and combined with the relevant knowledge of social networks and data mining,this thesis explores the community structure and the relationship between users in location-based social networks.The social activities of moving objects are studied from the micro level,and the evolution rules and influence mechanism of social relations of moving objects are explored from the macro level.The main research contents of this thesis include the following aspects:(1)Based on the check-in data,the evolution calculation method of social relationship among users in location-based social networks is studied.Social relationship is an important part of location social network,and its strength is highly related to the spatial and temporal information of location social network.Based on the check-in data,this thesis proposes an information entropy-based calculation method for the evolution of user social relationship.Firstly,by analyzing the spatial-temporal interactivity of the location social network,the spatial-temporal information of the check-in data is extracted to construct the user interaction vector.Secondly,the concept of information entropy is introduced to quantitatively represent the strength of users’ social relationship by the time diversity of the distribution of users’ interactive behaviors.Finally,the thesis deeply analyzes the changing characteristics of spatial-temporal interaction between users in the location social network,and combines the user interest drift algorithm to build a social relationship evolution model to calculate the evolution of user social relationship strength.Experiments show that this method can effectively improve the accuracy of calculating the strength of users’ social relationships.(2)Based on social activities,the structural evolution calculation method of location-based social network community is studied.Community structure is the basis of social network analysis.Based on social activities,this thesis studies the rule of community structure changing with time,events,people and space,and proposes a location-centric community structure evolution calculation method.Firstly,the time model of social network is constructed based on the spatial-temporal interaction of users.Secondly,location-centric communities are discovered by combining location and network topology.Finally,by extracting the communities in the continuous time window to analyze their structural changes,the evolutionary characteristics of the community structure in location-based social networks are comprehensively discovered.Experiments show that this method can effectively improve the computational efficiency and accuracy of community discovery and community structure evolution.(3)Based on the research results of this thesis,a prototype system of spatiotemporal social relationship evolution calculation is designed and implemented.In order to verify the effectiveness of the proposed algorithm in the real datasets,this thesis designs and develops a spatiotemporal social relationship evolution calculation prototype system on the basis of the research results include user social relationship strength evolution calculation based on information entropy and locationcentric community structure evolution calculation.From the perspective of practice,the evolution of moving objects in location-based social networks is explored.