Complex network anomaly recognition integrating multi-source and large data

ABSTRACT

With the continuous expansion of the network scale, the classical complex network recognition algorithm cannot efficiently deal with the existing large-scale network graph data, and proposes a complex network anomaly recognition method which combines multi-source and large data. First, the objects with the same semantic features are merged into the same abstract individual, and the finite individual description is used. An arbitrary number of network structures, based on the analysis of morphological analysis technology, identify the predicates in the three-valued logic structure of the network with a specific management operation extending Sagiv on the network to describe the semantic features of the abnormal recognition operation on the network. The method of changing abnormal behavior is identified. Finally, an example analysis in open-source software JEdit shows the effectiveness of the method.

KEYWORDS:

• Multi-source large data
• complex network
• anomaly recognition

Additional information

Notes on contributors

Xiangfei Meng

Xiangfei Meng, utilize modern Inner Mongolia vocational and technical college lecturer, professional leader, wulanhaote, China. Jilin university master's degree in computer science and technology, research field is computer application technology.