Simplified visual analysis method for large-scale multi-element network data

• Main Authors: ZHANG RUMIN, CHOU DAVID, HU MIAOXIN, LIU YUHUA, WANG YIGANG
• Format: Patent
• Language: Chinese; English
• Published: 14.09.2021
• Subjects: CALCULATING; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; HANDLING RECORD CARRIERS; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS

The invention discloses a simplified visual analysis method for large-scale multi-element network data. The method comprises the following steps: based on large-scale data of an original network, constructing an attribute-enhanced network representation learning model, and converting nodes into high-dimensional vector representation of an embedded topological structure and attribute information; then constructing a multi-level clustering model by using an attribute enhanced network representation learning model, and dividing nodes into level categories in a vectorization space according to structure compactness, attribute homogeneity and a clustering number; finally, designing a simplified expression visual analysis scheme, and constructing a simplified visual analysis system of the large-scale multi-element network data. According to the simplified visual analysis system, visual expression is formed through clustering views and collaborative views. According to the method, visual simplification, exploration