Method and system for grouping netizens based on historical behaviors and interestingness

Abstract

The invention provides a method and a system for performing group division on netizens based on historical behaviors and interestingness. The method comprises the following steps: selecting netizens,and extracting a first information publishing / forwarding record and a second information publishing / forwarding record corresponding to the netizens; determining an overlapping coefficient, a first influence coefficient and a second influence coefficient of the first interest point and the second interest point according to a relationship between the first information publishing / forwarding record and the second information publishing / forwarding record; determining a relationship strength coefficient between the first netizen and the second netizen according to the first influence coefficient, the second influence coefficient and the coincidence coefficient; and dividing the netizen pairs of which the relationship strength coefficients exceed a set threshold value into the same group. According to the scheme, netizens in the same group can have a relatively strong influence relationship with one another, and an accurate team division basis is provided for subsequent disambiguation processing between different group viewpoints.

Description

technical field [0001] The invention relates to the technical field of information processing, in particular to a method and system for grouping netizens based on historical behavior and degree of interest. Background technique [0002] The Internet has become inseparable from people's daily life, work and study, and network information contains a large amount of information generated by netizens. However, the meaning of the information generated by netizens cannot be determined solely by the literal meaning. It also contains rich emotional information, not only contains its explicit emotional information, but also contains a large number of implicit emotions, including irony, irony, negation, etc. How to mine the texts containing the hidden emotions of netizens from massive amounts of information and analyze their emotional tendencies is the prerequisite for the analysis of individual and group sentiments of netizens. On this basis, the purpose of event relationship disamb...

AI Technical Summary

Problems solved by technology

In the prior art, netizens are only divided into the same group according to the simple connection between them, such as those who register in the same forum are the same group, but in actual conditions, different netizens in the same forum do not have any relationship with each other. It is difficult to achieve the purpose of event relationship disambiguation

Images