Scientific research community discovery method and device based on label propagation

Abstract

The invention discloses a scientific research community discovery method and device based on label propagation. The community discovery method comprises the following steps: according to the similarity of attributes of nodes at two ends of edges in an initial network, adjusting weights of all the edges in the network to obtain a new network; after a new network is initialized, traversing all edges in the network for multiple times according to the topological structure of the network and the weights of the edges to carry out multi-label propagation, and when labels of all nodes in the network do not change any more, ending the multi-label propagation process to obtain a final label list of all the nodes; and according to the final label list of all the nodes in the network, dividing the network into a preset number of communities to discover overlapping communities. According to the community discovery method, the nodes in the finally divided communities are tightly connected and have high node attribute similarity, the nodes among the communities are sparsely connected and have low attribute similarity, and overlapped communities can be discovered. In addition, the community discovery method is low in time complexity and can be applied to a large-scale network.

Description

technical field [0001] The invention relates to a method for discovering a scientific research community based on tag propagation, and also relates to a corresponding device for discovering a scientific research community, belonging to the technical field of complex network analysis. Background technique [0002] A complex network is a network defined by scholars, which is characterized by a large scale, often with more than 100,000 nodes, and complex connection relationships, which contain rich information. Social networks, biological networks, and academic networks are all typical complex networks, so research on complex networks is of great significance. Community detection method is a kind of method commonly used in the study of complex networks; such as figure 1 As shown, by dividing the complex network into multiple communities, the internal connections of the communities are tight, and the connections between the communities are sparse, so that the community structur...

AI Technical Summary

Problems solved by technology

[0004] However, most of the current community discovery methods only study the topology of the network, but do not pay attention to the node attribute information that also contains rich information; in the community discovery methods that consider the node attribute information, it is common that overlapping communities cannot be found, and it is difficult to apply to large-scale scale network issues

Images