User behavior driven influence maximization method in evolutionary social network

Abstract

The invention discloses a user behavior-driven influence maximization method in an evolved social network, which comprises the following steps of: 1, acquiring behavior data of a user in the social network, and establishing an evolved social network model evolved along with time; step 2, according to the behavior logs of the users, calculating an information spreading probability between the users to obtain weights of edges in the evolutionary social network; 3, improving the independent cascade propagation model, and establishing an independent cascade propagation model driven by user behaviors; step 4, according to an improved reverse sampling method, querying k nodes which have the most influence in the evolved social network at different time points, namely a seed node set; and 5, verifying the validity of the independent cascade propagation model driven by the user behaviors. The problem that the user with the largest influence cannot be accurately inquired on the evolutionary social network due to the fact that the influence of user behaviors on information spreading is not considered in the prior art can be solved.

Description

technical field [0001] The invention relates to the technical field of social networks, in particular to a user behavior-driven influence maximization method in an evolutionary social network. Background technique [0002] With the development of Internet technology, more and more virtual social networks appear one after another, such as the large-scale social networking site Facebook, and the interpersonal network formed by mobile communication. Compared with traditional social networks, these emerging social networks have strong dynamic characteristics and are called Evolving Social Networks. An evolutionary social network is a network that changes over time. During the entire evolution process, the number of users in the network and the relationship between users will continue to change, and the influence between users will not remain constant. , but will change with the change of user behavior, so it is more reasonable to focus on evolutionary characteristics to model s...

AI Technical Summary

Problems solved by technology

[0010] For this reason, the purpose of the present invention is to propose a method for maximizing influence driven by user behavior in an evolutionary social network to solve the problem that the existing technology does not consider the influence of user behavior on information dissemination, and cannot accurately query the influence on the evolutionary social network. The most powerful user problem

Images