Enhanced Community Structure Detection in Complex Networks with Partial Background Information

Zhang, Zhongyuan; Sun, Kai-Di; Wang, Siqi

doi:10.1038/srep03241

Cited by 53 publications

(27 citation statements)

References 14 publications

(36 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The most widely-used approach has been to employ pairwise constraints, either must-link or cannot-link, which indicate that either two nodes must be in the same community or must be in different communities. This strategy has been implemented via several algorithms, including modularity-based methods (Li et al 2014), spectral partitioning methods (Habashi et al 2016;Zhang 2013;Zhang et al 2013), a spin-glass model (Eaton and Mansbach 2012), and matrix factorization methods (Shi et al 2015;Zhang 2013). Such approaches have often provided significantly better results on benchmark data, when compared to standard unsupervised algorithms.…”

Section: Semi-supervised Learning In Community Findingmentioning

confidence: 99%

Active semi-supervised overlapping community finding with pairwise constraints

Alghamdi

Greene

2019

Appl Netw Sci

View full text Add to dashboard Cite

Algorithms for finding communities in complex networks are generally unsupervised, relying solely on the structure of the network. However, these methods can often fail to uncover meaningful groupings that reflect the underlying communities in the data, particularly when they are highly overlapping. One way to improve these algorithms is by incorporating human expertise or background knowledge in the form of pairwise constraints to direct the community detection process. In this work, we explore the potential of semi-supervised strategies to improve algorithms for finding overlapping communities in networks. We propose a method, based on label propagation, for finding communities using pairwise constraints. Furthermore, we introduce a new strategy, inspired by active learning, for intelligent constraint selection, which is designed to minimize the level of human annotation required. Extensive evaluations on synthetic and real-world datasets demonstrate the potential of this strategy for effectively uncovering meaningful overlapping community structures, using a limited amount of supervision.

show abstract

Section: Semi-supervised Learning In Community Findingmentioning

confidence: 99%

Active semi-supervised overlapping community finding with pairwise constraints

Alghamdi

Greene

2019

Appl Netw Sci

View full text Add to dashboard Cite

show abstract

“…Recently, many semi-supervised community detection algorithms have been proposed [6]- [9]. Ma el al proposed a semi-supervised method based on symmetric non-negative matrix, which incorporates the pairwise constraints into the adjacency matrix for finding the community structure [6].…”

Section: Introductionmentioning

confidence: 99%

“…A semi-supervised method based on the spin-glass model from statistical physics can integrate the prior information in forms of individual labels and pairwise constrains into community detection proposed by Eaton and Mansbach [7].Zhang [8] studied a semi-supervised learning framework which encodes pairwise constraints by modifying the adjacency matrix of network, which can also be regarded as de-noising the consensus matrix of community structures. Later, Zhang [9] added a logical inference step to utilize the must-link and cannot-link constraints fully. These algorithms use the prior information by transferring and modifying the adjacency matrix directly.…”

Section: Introductionmentioning

confidence: 99%

Semi-supervised Community Detection Framework Based on Non-negative Factorization Using Individual Labels

Wang

Xue

et al. 2015

Advances in Swarm and Computational Intelligence

View full text Add to dashboard Cite

Abstract. Community structure is one of the most significant properties of complex networks and is a foundational concept in exploring and analyzing networks. Researchers have concentrated partially on the topology information for community detection before, ignoring the prior information of the complex networks. However, background information can be obtained from the domain knowledge in many applications in advance. Especially, the labels of some nodes are already known, which indicates that a point exactly belongs to a specific category or does not belong to a certain one. Then, how to encode these individual labels into community detection becomes a challenging and interesting problem. In this paper, we present a semi-supervised framework based on non-negative matrix factorization, which can effectively incorporate the individual labels into the process of community detection. Promising experimental results on synthetic and real networks are provided to improve the accuracy of community detection.

show abstract

“…Zhang et al extend this framework to other methods including modularity maximization model and Infomap algorithm [ 9 ]. Then Zhang et al further extend this framework by adding a logical inference step to better utilize the supervised information [ 10 ]. This kind of methods often ignore the difference between the pairwise relationship in the network topology and pairwise constraint.…”

Section: Introductionmentioning

confidence: 99%

Exploring the roles of cannot-link constraint in community detection via Multi-variance Mixed Gaussian Generative Model

Yang

Jin

et al. 2017

PLoS ONE

View full text Add to dashboard Cite

Due to the demand for performance improvement and the existence of prior information, semi-supervised community detection with pairwise constraints becomes a hot topic. Most existing methods have been successfully encoding the must-link constraints, but neglect the opposite ones, i.e., the cannot-link constraints, which can force the exclusion between nodes. In this paper, we are interested in understanding the role of cannot-link constraints and effectively encoding pairwise constraints. Towards these goals, we define an integral generative process jointly considering the network topology, must-link and cannot-link constraints. We propose to characterize this process as a Multi-variance Mixed Gaussian Generative (MMGG) Model to address diverse degrees of confidences that exist in network topology and pairwise constraints and formulate it as a weighted nonnegative matrix factorization problem. The experiments on artificial and real-world networks not only illustrate the superiority of our proposed MMGG, but also, most importantly, reveal the roles of pairwise constraints. That is, though the must-link is more important than cannot-link when either of them is available, both must-link and cannot-link are equally important when both of them are available. To the best of our knowledge, this is the first work on discovering and exploring the importance of cannot-link constraints in semi-supervised community detection.

show abstract

Enhanced Community Structure Detection in Complex Networks with Partial Background Information

Cited by 53 publications

References 14 publications

Active semi-supervised overlapping community finding with pairwise constraints

Active semi-supervised overlapping community finding with pairwise constraints

Semi-supervised Community Detection Framework Based on Non-negative Factorization Using Individual Labels

Exploring the roles of cannot-link constraint in community detection via Multi-variance Mixed Gaussian Generative Model

Contact Info

Product

Resources

About