A spectral clustering-based framework for detecting community structures in complex networks

Jiang, Jeffrey Q.; Dress, Andreas; Yang, Genke

doi:10.1016/j.aml.2009.02.005

Cited by 39 publications

(33 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We managed to reduce the effect by visually inspecting each sub-network with more than 100 phenotypes in the first level modules while automatically decomposing the phenotype network using our previous algorithm [12].…”

Section: Extracting the Modules Of The Phenotype Networkmentioning

confidence: 99%

Towards Prediction and Prioritization of disease genes by the modularity of human phenome-genome assembled network

Jiang

Dress²,

Chen

2010

Journal of Integrative Bioinformatics

Self Cite

View full text Add to dashboard Cite

SummaryEmpirical clinical studies on the human interactome and phenome not only illustrates prevalent phenotypic overlap and genetic overlap between diseases, but also reveals a modular organization of the genetic landscape of human disease, provding new opportunities to reduce the complexity in dissecting the phenotype-genotype association. We here introduce a network-module based method towards phenotype-genotype association inference and disease gene identification. This approach incorporates protein-protein interaction network, phenotype similarity network and known phenotype-genotype associations into an assembled network. We then decomposes the resulted network into modules (or communities) wherein we identified and prioritized the disease genes from the candidates within the loci associated with the query disease using a linear regression model and concordance score. For the known phenotype-gene associations in the OMIM database, we used the leave-one-out validation to evaluate the feasibility of our method, and successfully ranked known disease genes at top 1 in 887 out of 1807 cases. Moreover, applying this approach on 850 OMIM loci characterized by an unknown molecular basis, we propose high-probability candidates for 81 genetic diseases.

show abstract

Section: Extracting the Modules Of The Phenotype Networkmentioning

confidence: 99%

Towards Prediction and Prioritization of disease genes by the modularity of human phenome-genome assembled network

Jiang

Dress²,

Chen

2010

Journal of Integrative Bioinformatics

Self Cite

View full text Add to dashboard Cite

show abstract

“…These methods assign the nodes to communities with different degree of belonging values and form overlapping communities. The most popular fuzzy center-based clustering model is fuzzy c-means (FCM) clustering and is mostly used in combination with other techniques for community detection (Jiang et al, 2009;Liu, 2010;Zhang et al, 2007). FCM clustering is the most well-known fuzzy clustering algorithm proposed by Dunn (1974) and extended by Bezdek (1981).…”

Section: Overlapping Community Detection Modelsmentioning

confidence: 99%

Fuzzy duocentric community detection model in social networks

2015

View full text Add to dashboard Cite

a b s t r a c tThe main goal of this paper is to present a clustering model to identify duocentric communities in the complex networks. A duocentric community is built around two central nodes which are as close as possible to other nodes, while the central nodes are connected enough to each other to shape the center of the community. To detect such communities, we develop a new objective function based clustering model. The network's nodes are assigned to the duocentric communities by the type-2 fuzzy numbers which indicate the degrees of belonging to the communities by upper and lower membership values. Generated interval type-2 fuzzy membership values by our proposed model are able to determine how much each node belongs to the both central nodes and how it is shared among communities. Also, the compatible verification index with the proposed model is introduced to evaluate and compare the results of the proposed model with the existing approach in the literature. Finally, the performance of the proposed algorithm is validated by detecting duocentric communities in three artificial networks and two real social networks.

show abstract

“…Many traditional community detecting methods hold that each node can only belong to one community, such as Modularity optimization [1], [2], Hierarchical clustering [3], [4], Spectral Algorithms [5], [6], label propagation algorithm [7], [8], Methods based on statistical inference [9]. However in some real networks, communities are not independent , nodes can belong to more than one community ,which will lead to overlapping communities .…”

Section: Relate Workmentioning

confidence: 99%

A Fast Method of Detecting Overlapping Community in Network Based on LFM

Li¹,

Zhu²

2015

JSW

View full text Add to dashboard Cite

Detect overlapping communities efficiently and effectively in various social networks has been more and more important. Aiming at the high complexity of expanding strategy and the defect of generating many homeless nodes in LFM, we propose a quick algorithm based on local optimization of a fitness function(QLFM). The proposed algorithm firstly select a node as seed randomly .With a local fitness function ,the algorithm then will expand from inside to outside of the seed according to the Breadth-First-Search in graph. As different seeds will expand to different communities independently ,and these communities have same nodes ,thus our method can detect overlapping nodes quickly and efficiently. An empirical evaluation of the method using real and synthetic datasets shows that the method give better result not only in time efficiency, but also in quality aspect than other methods at the overlapping community detection.

show abstract

A spectral clustering-based framework for detecting community structures in complex networks

Cited by 39 publications

References 12 publications

Towards Prediction and Prioritization of disease genes by the modularity of human phenome-genome assembled network

Towards Prediction and Prioritization of disease genes by the modularity of human phenome-genome assembled network

Fuzzy duocentric community detection model in social networks

A Fast Method of Detecting Overlapping Community in Network Based on LFM

Contact Info

Product

Resources

About