Attribute graph literature clustering method based on graph convolutional neural network

Abstract

The invention discloses an attribute graph literature clustering method based on a graph convolutional neural network, and belongs to the field of graph data mining. Specifically, literature attribute graph feature learning is carried out by using a cross-layer linked graph convolutional neural network; estimating an optimal cluster number from the node features by using a deep clustering estimation model; alternately executing the two steps to complete training; utilizing the trained model to obtain the characteristics of all to-be-clustered literature attribute graph nodes and the estimated number of clustering clusters; and taking the characteristics and the estimated number of the clustering clusters as input, and obtaining a clustering result of the literature attribute graph by using a k-means clustering method. When a cross-layer linked graph convolutional neural network is trained, a self-separation regularization item based on node pairwise similarity is adopted, so that the characteristics of nodes in the same cluster are similar and the characteristics of nodes in different clusters are far away, and the performance of graph clustering is effectively improved. And the clustering estimation module realizes data-driven clustering cluster number estimation, so that the whole system is more suitable for a real data environment without labels.

Description

technical field [0001] The invention belongs to the field of graph data mining, and in particular relates to a graph convolutional neural network-based attribute graph document clustering method. Background technique [0002] Attribute graph clustering is a basic task in the field of graph data mining. Its purpose is to divide the nodes in the graph into mutually disjoint clusters according to node attributes and graph structure information. Compared with traditional graph clustering methods that only use graph structure information, attribute graph clustering is more suitable for scenarios where nodes have rich content information. Attribute graph clustering has a wide range of practical applications in the fields of community discovery, protein functional module detection, and financial network fraud detection. [0003] A large number of graph clustering works based on deep models have been proposed. Compared with shallow graph clustering methods, deep methods are better...

AI Technical Summary

Problems solved by technology

[0006] Aiming at the problems existing in the above-mentioned prior art, the present invention proposes an attribute graph document clustering method based on a graph convolutional neural network, which is used to solve the problem of lack of utilization of document citation relations in the process of document clustering, and it can deal with From the unbalanced cluster structure in real graph data, learn node features that are friendly to clustering tasks, and estimate the number of clusters in graph data according to node features, and realize parameter-free attribute graph clustering

Images