Contact semantic grouping method for network message communication

Abstract

The invention discloses a contact semantic grouping method for message communication, which is mainly used for solving the problem of inconvenient use of a social network communication system because the contact semantic grouping is difficultly and automatically realized in a large-scale social network message communication system in the prior art. The method comprises the following steps of: constructing a communication social relation picture of users by using historical information of message communication; then selecting nodes correspondingly frequently communicated with nodes of the users as semantic groups, finding out all candidate semantic groups from the semantic groups, and constructing topological relation formed from the semantic groups by removing meaningless semantic groups and combining irrelevant semantic groups; and finally, constructing layered semantic groups by calculating Jaccard coefficients. The method disclosed by the invention is capable of processing large-scale social network message communication data and has the advantages of being simple and easy, higher in accuracy and rapider in operation speed. The method can be used for solving buddy grouping problems of mailboxes, instant messaging and social network communication.

Description

technical field [0001] The invention belongs to the technical field of network communication, relates to network user communication data and behavior analysis technology, in particular to a semantic grouping method for contacts used in network message communication. For network message communication systems such as web mail, mobile phone calls, instant messaging, and social networking, it provides automatic group management support for user contacts. Background technique [0002] With the advent of the information age, many fields are currently facing massive data processing, and how to extract useful information from massive data has become the main direction of current research. Especially as there are more and more personal communication methods, their historical communication data widely exist on the clients of network communication systems such as mobile phones, emails, instant messages, Fetion, social networks, etc., and the roles of individuals in society are diverse....

AI Technical Summary

Problems solved by technology

However, these traditional graph clustering methods usually use a global batch data processing method for large-scale data sets, resulting in low efficiency, long running time and large storage space overhead. For the historical message communication data of a single user, The semantic grouping of contacts processed by it is difficult to meet the practical requirements

Images