Node Attribute-enhanced Community Detection in Complex Networks

Abstract

Community detection involves grouping the nodes of a network such that nodes in the same community are more densely connected to each other than to the rest of the network. Previous studies have focused mainly on identifying communities in networks using node connectivity. However, each node in a network may be associated with many attributes. Identifying communities in networks combining node attributes has become increasingly popular in recent years. Most existing methods operate on networks with attributes of binary, categorical, or numerical type only. In this study, we introduce kNN-enhance, a simple and flexible community detection approach that uses node attribute enhancement. This approach adds the k Nearest Neighbor (kNN) graph of node attributes to alleviate the sparsity and the noise effect of an original network, thereby strengthening the community structure in the network. We use two testing algorithms, kNN-nearest and kNN-Kmeans, to partition the newly generated, attribute-enhanced graph. Our analyses of synthetic and real world networks have shown that the proposed algorithms achieve better performance compared to existing state-of-the-art algorithms. Further, the algorithms are able to deal with networks containing different combinations of binary, categorical, or numerical attributes and could be easily extended to the analysis of massive networks.

A Distributed Bagging Ensemble Methodology for Community Prediction in Social Networks

“…Due to the profitable application of community detection in plenteous scientific areas, an affluent set of algorithms [2][3][4][5][6][7][8][9][10] has already been published to tackle this a NP-hard class problem. From methods that are exclusively based on the repetitive calculation of a global network topology metric, to alternatives inspired by discrete mathematics and physics, the pluralism of classic community detection processes is indeed remarkable.…”

“…The iterative processes aiming to identify and remove all the inter-connection edges, by recursively maximizing a global topology criterion are classified as divisive algorithms [2,3,[6][7][8]. In these techniques, at each iteration step, a finer community hierarchy layer is formed.…”

A generative model for exploring structure regularities in attributed networks

“…Obviously, using only one type of information will ignore another type of information. It has shown that combing network topology with attribute information can not only improve the quality of community detection, but also has potential to provide the semantic descriptions of communities, and help to understand the functions of communities [7][8][9][10][11].Existing methods that joint the two types of information can be roughly classified into two categories: model-based methods [12][13][14][15][16][17][18][19][20][21] and other heuristic methods [22][23][24][25][26][27]. Model-based methods are mainly on the basis of probabilistic generative models.…”

“…Obviously, using only one type of information will ignore another type of information. It has shown that combing network topology with attribute information can not only improve the quality of community detection, but also has potential to provide the semantic descriptions of communities, and help to understand the functions of communities [7][8][9][10][11].…”

Community Detection in Attributed Social Networks: A Unified Weight-Based Model and Its Regimes

“…[43,75,146,180,224] and many other papers cited in this survey, topology and semantics often provide complementary information and thus combining them usually leads to achieving better performance in community detection. For example, the semantics may compensate the sparseness of a real network [106]. At the same time, topological information may be helpful if there are missing or noisy attributes [180].…”