A method and device for acquiring social relationship types of network subjects

Abstract

Social network behavior data of two network subjects whose social relationship type is required to be determined is obtained. The network social behavior data of the two network subjects to be determined is discretized by a preset manner in such a way that each of the social behavior is discretized into N levels according to a quantity of each social network behavior. A merged probability of each social relationship type to which the two network subjects belong is calculated according to a conditional probability that each level of network social behavior corresponds to each social relationship type. A social relationship type having the largest merged probability is initially determined as a social relationship type of the two network subjects. The present techniques effectively utilized network social behavior data, implement determination of social relationship type, and improve user network experiences.

Description

technical field [0001] The invention relates to network technology, in particular to a method and device for acquiring social relationship types of network subjects in the network. Background technique [0002] With the development of the Internet, the Internet has developed into a comprehensive information platform integrating communication, information, entertainment, search, e-commerce, office collaboration and enterprise customer service. Due to the diversification of functions, the relationship between users needs to be established on this platform. [0003] In the prior art, each web site has its own user group, the relationship between users can form a network, and the dynamic relationship between users can form a relationship model. When establishing a user relationship model for these users, some algorithms are usually used to perform matching calculations on users. According to the results of matching calculations, for potential users with high mutual relationship...

AI Technical Summary

Problems solved by technology

[0004] However, with the development of Internet applications, the interaction between network users becomes more frequent and the interaction methods become more diversified. Usually, a user will have dozens or even hundreds of contacts, and there are several contacts between each contact. There are different types of social relationships, and in this case, the way of establishing user relationships using existing technologies is limited by the voluntary degree of users to provide mutual social relationship information, and cannot fully grasp the actual social relationships between users; in addition, Designing an overly complex relationship classification option for user confirmation will lead to a decrease in the participation of network subjects. Therefore, the social relationship division of network subjects in the prior art is not very rich, and cannot reflect the complexity of the actual user relationship types. The above situation, Make e-commerce manufacturers unable to fully grasp the complex interrelationship information among various network entities, and reduce the accuracy of various e-commerce operations

Another major defect of the existing technology is that the efficiency of social relationship confirmation through the voluntary selection of users is too low

In short, the above situation not only affects the search of target customers for e-commerce based on social networks, but also hinders users' experience of network service products.

Images