Abstract:With the widespread adoption of the Internet and an increase in the number of Internet users,the security of the network environment has been seriously threatened by Internet public hazards organizations,such as the In-ternet water army and the internet public hazards community.Given the characteristics of these organizations,which include being distant from domestic supervision and possessing strong concealment and camouflage abili-ties,a social network characteristic analysis method was proposed for Internet public hazards organizations.Ini-tially,at the user characteristic level,the influence distribution of group members was calculated through centrality measurement to identify the core users with the greatest influence,and an analysis of the anonymity characteristics of these core members was conducted.Subsequently,at the complex network characteristic level,the cluster status and resource distribution within the community were studied through the analysis of small-world effects and scale-free characteristics.Finally,at the temporal characteristic level,after preprocessing the post content of core mem-bers,BERTopic was utilized for dynamic topic modeling to ascertain the relationship between community themes and time.A large number of experimental performance analyses have verified the effectiveness of the proposed method.The research results can be utilized to analyze the characteristics of public hazard communities,track criminal activity trajectories,and provide theoretical references for the investigation and crackdown of public haz-ard organizations,as well as for the design of automated detection systems.

Modeling method of internet public information data mining based on probabilistic topic model

A Knowledge-Based Semisupervised Hierarchical Online Topic Detection Framework.

Research on data mining models for the internet of things

A Intelligent Method of Modelling Web User Interest

A Pattern of Experiment Data Mining Based on Π Theorem Combined with ANN

Probabilistic model for academic social network and its applications

A Survey on News Text Visualization via Probabilistic Topic Modeling

Interactive Topic Modeling Based on Hierarchical Dirichlet Process

Hierarchical Community Discovery for Social Networks Based on Probabilistic Topic Model

Analysis Method on the Properties of Social Networks of Internet Nuisance Communities

Research on Principle Techniques for Network Intrusion Detection Based on Data Mining and Analysis Approach

MIT: Mutual Information Topic Model for Diverse Topic Extraction

Security Analysis of Social Network Topic Mining Using Big Data and Optimized Deep Convolutional Neural Network

The Design of Police Big Data Mining Model

Study on application of open source intelligence from social media in the military

Research on the Hidden Markov Model in Hot Topics of the Internet

Research on Mining Common Concern Via Infinite Topic Modelling

ACT-LDA: A Probabilistic Model of Topic, Community and User Influence

Investigative Data Mining For Counterterrorism

Simulation of Computer Network Information Security Assessment Model Based on Data Mining

Heterogeneous Latent Topic Discovery for Semantic Text Mining