Microblogging network user behavior analyzing method based on MMSB

Abstract

The invention discloses a microblogging network user behavior analyzing method based on MMSB. The method includes setting mixed membership and a transition probability matrix B of a simulated microblogging network to be established, and establishing an MMSB simulated microblogging network; acquiring a joint probability formula of microblogging forwarding percentage p (Y) between two optional nodes in the entire microblogging network through a model according to the MMSB simulated microblogging network established in the step 1 and the acquired mixed membership and transition probability matrix B; analyzing by a mean-field variable method according to the expression of the microblogging forwarding percentage p (Y) acquired in the step 2; figuring out actual mixed membership and a transition probability matrix B of a microblogging network by the coordinate ascent algorithm according to the user's microblogging forwarding percentage p (Y) actually measured in the simulated microblogging network and the inequality acquired in the step 3. The method is used for analyzing behavior information of a microblogging network user, and is capable of obtaining more actual user's information.

Description

technical field [0001] The invention belongs to the technical field of network applications, and in particular relates to the design of an MMSB-based microblog network user behavior analysis method. Background technique [0002] Weibo is an online social network, which provides a network communication platform for registered users, through which users can conduct a series of social activities, such as business, making friends, obtaining information, interacting with friends, etc. Weibo has developed into A very important information dissemination and communication tool has become a kind of enrichment and expansion of real social interaction. Weibo has a large number of registered users, and each user has different behaviors. The difference in user behavior is mainly reflected in each user's interests, hobbies, content of concern, frequency of posting Weibo, and frequency of forwarding Weibo. Therefore, analyzing user behavior is of great practical significance to the entire...

AI Technical Summary

Problems solved by technology

Although the MMBS user behavior analysis method takes into account the correlation between data and the multi-role of users, it cannot be directly applied to the microblog network for user behavior analysis, because the MMSB model divides the connection relationship between users into connection and The two relationships are not connected, and it does not take into account the fact that users in the microblog network forward multiple microblogs, that is, there are multiple connections between nodes

[0004] For Weibo, the existing user behavior analysis methods have certain limitations: many current user behavior analysis methods obtain data from a single object, and assume that the data in the network is independent, or independent under certain conditions. However, the current microblog user data are all related, so the results obtained by using these methods to analyze the microblog network are inaccurate; the existing user behavior analysis methods assume that the user only belongs to one community, and cannot mine users across The multi-role information of the community; the MMSB user behavior analysis method takes into account the data relevance, and can also mine the multi-role information of the user, but only divides the connection relationship of users in the network into two situations: connection and non-connection, and does not take into account microblog Connection weight between users

Therefore, the MMSB user behavior analysis method is not suitable for Weibo reposting network

Images