Visual analysis method based on online social media personal center network

Abstract

The invention discloses a visual analysis method based on an online social media personal center network, belongs to the technical field of electronic information, and solves the problem that the effectiveness of recommendation cannot be ensured in the prior art. The method includes: on the basis of the online friend relationship data and offline sign-in data of each central user, calculating thecommunity characteristics and the distribution of repeated sign-in POI types; constructing social feature parameters of the consistency of the online relation and the offline activity, and selecting features capable of reflecting social behavior characteristics of all aspects from the social feature parameters to form social feature vectors; dividing friend types different social circle friend types by adopting an unsupervised clustering algorithm based on the social feature vectors, namely a POI type social network; performing visual design coding based on the community characteristics and the distribution of the repeated sign-in POI type to obtain a corresponding visual view; based on the POI type social network and the visual view, achieveing visual analysis of the social network from the overall social network to the POI type social network to the central user level. The method is used for visual analysis of the personal center network.

Description

technical field [0001] A visual analysis method based on an online social media personal-centric network belongs to the field of electronic information technology and is used for visual analysis of a personal-centric network. Background technique [0002] Online social network has replaced the traditional social mode and has become one of the main ways for people to socialize. Based on social service platforms, the scale of information disseminated on social networks is unprecedented. This information is stored by the social service platform as social data for further user behavior research, assisting managers in decision-making or academic research. The research value of these data is huge. The massiveness, heterogeneity, and multidimensionality of social data have brought great challenges to its research. Traditional social data research methods such as statistical methods and data mining methods are more or less There are few limitations. As a new branch of data visual...

AI Technical Summary

Problems solved by technology

[0009] In view of the problems of the above research, the purpose of the present invention is to provide a visual analysis method based on the online social media personal center network, to solve the problem in the prior art that the recommendation based on friend relationship data and location check-in data does not carry out the feature of friends. Community structure processing and the division of community functions without considering the POI type of the check-in location, so that social types cannot be divided, and more personalized recommendation algorithm design cannot be customized according to different social types; and the line of community structure cannot be quickly discovered. The POI type of the next activity, so that the recommendation impact factor between the check-in positions of different social users cannot be calculated based on the POI type of the social user's circle of friends, so the effectiveness of the recommendation cannot be guaranteed

Images