Microblog water army detection method and system based on artificial immune danger theory

Abstract

The invention belongs to the technical field of microblog networks, and discloses a microblog water army detection method and system based on an artificial immunization danger theory. The method is characterized by applying an artificial immunization thought to the detection of the microblog user behavior characteristics and obtaining the microblog user data by using a focused web crawler; describing and defining network water army behaviors through an analysis method based on user behavior characteristics, and distinguishing characteristic attributes of novel network water army and normal users; and finally, applying a signal processing mechanism of an artificial immune danger theory to network water army detection, and detecting water army users in microblogs by adopting a dendritic cellalgorithm DCA of the danger theory. According to the method, the microblog user data is acquired in a Python-based focusing web crawler mode and stored in the database in a structured data mode, a data set is easily acquired in the mode, various behavior data of a user can be reasonably acquired, and the method has the advantages of being short in crawling period, high in data quality and the like.

Description

technical field [0001] The invention belongs to the technical field of microblog network, and in particular relates to a method and system for detecting microblog trolls based on artificial immune risk theory. Background technique [0002] At present, the existing technologies commonly used in the industry are as follows: [0003] Microblogging trolls refer to those who are motivated by profit, in order to achieve the purpose of confusing the authenticity of information, guiding wrong public opinion, and harming the interests of citizens, by manipulating software robots or troll accounts to create and spread falsehoods on Weibo. Generic term for spam generators such as opinion and spam. Weibo troll detection uses some data mining techniques to define highly distinguishable features or behavior patterns to discover hidden cyber trolls. [0004] At present, the main water army detection methods are as follows: [0005] Trouble detection method based on content features: inc...

AI Technical Summary

Problems solved by technology

[0014] (2) The traditional water army detection method based on environmental features has great difficulties in data acquisition, and the generalizability is low

[0015] (3) At present, the behavior of the water army is becoming more and more complicated. Selecting local and single features for the detection of the water army will have incomplete feature descriptions, which will easily lead to problems such as recognition errors.

[0016] (4) The traditional classification detection method based on user characteristics requires a large number of training data sets, and the detection efficiency is low and the applicability is not high

Images