Personalized social network influence identifying method

Abstract

The invention discloses a personalized social network influence identifying method and relates to the social engineering science. The personalized social network influence identifying method aims at identifying the personalized influence relation in the network according to the interpersonal interaction in the network. The personalized social network influence identifying method comprises the technical points of selecting a network user group, setting up relation pairs of a targeted user relative to direct friends of the targeted user, extracting the relation characteristic value of each relation pair, determining the influence extent of each relation pair, selecting a predicting model, training the predicting model by using the relation characteristic value and the influence extent of each relation pair, selecting a new user, setting up the relation pairs of the new user relative to the direct friends of the new user, extracting the relation characteristic value of each relation pair, substituting the relation characteristic value of each relation pair of the new user into the trained predicting model, and calculating the influence extent of each direct friend of the new user on the new user by the trained predicting model.

Description

technical field [0001] The present invention relates to social engineering, especially a method for searching and discovering personalized influencers of target users in a local area network by using social network analysis (Social Network Analysis) and machine learning (Machine Learning) technologies. Background technique [0002] With the continuous development of the network, it is generally believed that there are many different people playing many roles around everyone in the social network, such as followers, leaders, close friends, ordinary friends, etc. People have different influences or degrees of influence. [0003] The importance of identifying the degree of influence between people is not only that this information can help us better understand the function of social networks and how information flows, but also that this information contains great value for marketing and product promotion. [0004] For example, people often trust recommendations from third part...

AI Technical Summary

Problems solved by technology

[0006] However, in the above traditional graph-based network analysis, all the individual and relational weights are considered to be the same: it does not reflect the difference between the target user's trusted friends or strangers.

Images