Adaptive community detection in complex networks using genetic algorithms

Guerrero, Manuel; Montoya, Francisco G.; Alcayde, Alfredo; Gil, C.

doi:10.1016/j.neucom.2017.05.029

Cited by 111 publications

(46 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Gurrero et al [19] proposed a generational genetic algorithm, named GGA+, that includes also population initialisation while the space of solutions is searched under the guidance of network modularity as in Shang et al [18]. Adaptive analysis of the characteristics of a network from different levels of detail according to analysts' needs is also supported.…”

Section: Background and Related Workmentioning

confidence: 99%

The Central Community of Twitter ego-Networks as a Means for Fake Influencer Detection

Tsapatsoulis

Anastasopoulou

Ntalianis

2019

2019 IEEE Intl Conf on Dependable, Autonomic and Secure Computing, Intl Conf on Pervasive Intelligence and Computing, Intl Conf

View full text Add to dashboard Cite

The central community of social networks, usually represented through the highest degree k-core of the corresponding graph, is proposed here as a compact representation of large social networks. We show that the central community of egocentric social media networks, such as the ego networks on Twitter and Instagram, tell us much more about the actual influence of the ego than the whole egocentric network itself. We also propose a novel genetic algorithm for the identification of central community of egocentric social networks and we examine the importance of the proper initialisation of this algorithm. The actual Twitter ego networks we used in our experiments along with the corresponding Python code are made publicly available for anyone who wishes to use them.

show abstract

Section: Background and Related Workmentioning

confidence: 99%

The Central Community of Twitter ego-Networks as a Means for Fake Influencer Detection

Tsapatsoulis

Anastasopoulou

Ntalianis

2019

2019 IEEE Intl Conf on Dependable, Autonomic and Secure Computing, Intl Conf on Pervasive Intelligence and Computing, Intl Conf

View full text Add to dashboard Cite

show abstract

“…After the clustering is completed, the community structure is divided by finding a local maximum module value [7][8]; In 2009, Dress discovered that unstable social structures in complex networks can be adequately detected by relaxing the modularity function (Q function) of Newman's algorithm [9]; in 2012, Bennett L proposed a hybrid integer non-linear programming (MINLP) based on the Newman algorithm for the classification of weighted networks and the detection of overlapping communities [11]; In 2013, Meo put forward an algorithm to enhance the discovery of existing communities by using network weighting strategy. The algorithm enhances the community discovery by adding the edge center weight to the network topology [10]; In 2014, Li put forward a vertex similarity probability (VSP) model, which can find the community structure when the type of complex network is unknown [12]; Ferreira presented a complex network community detection algorithm based on time series clustering in 2016 [14]; In 2017, Guerrero proposed a complex network adaptive community detection algorithm GGA based on genetic algorithm, which is guided by the modular index [14]; Soundarajan et al put forward a HICODE meta-method in 2017 to find hidden communities that could find existing hidden communities [15]; Kumar et al developed a complex network community detection algorithm based on rough set, which constructed the rough set using the regional connectivity around the nodes [16]. In the study of bid rigging of water projects, Newman's fast algorithm was chosen to divide the complex network community structure because the community structure of the complex network composed of bidding enterprises was undiscovered.…”

Section: A Literature Review Of Community Detection Algorithms In Comentioning

confidence: 99%

The Analysis of Water Project Bid Rigging Behavior Based on Complex Network

Cheng¹,

Liu²,

Meng³

et al. 2017

Proceedings of the 2017 International Conference on Applied Mathematics, Modeling and Simulation (AMMS 2017)

View full text Add to dashboard Cite

Abstract-There are a series of issues in the development of water project bidding areas, the illegal act of bid rigging and colluding is particularly worth attention, and how to prevent the rigging and colluding behavior in the bidding process has already become a current hot topic. The community structure characteristics of complex networks was applied to analyze the bid rigging and colluding behavior of tenderers, the community detection model of bid rigging and colluding based on complex network was established, and the rationality and feasibility of the model were illustrated by analyzing the winning rate of the tenderers in the community, which showed that the community detection model of bid rigging and colluding based on complex network could provide some reference for the bidding supervision departments to identify the rigging colluding behavior.

show abstract

“…This iterative strategy improves the ability to identify small clusters. Community detection in a complex network is a variant of the clustering concept [14][15][16]. Blondel et al [17] tested the high accuracy of the Louvain algorithm on ad hoc modular networks and demonstrated its excellent performance in comparison with other community detection methods.…”

Section: Introductionmentioning

confidence: 99%

Dimension Reduction and Clustering Models for Single-Cell RNA Sequencing Data: A Comparative Study

Feng

Liu

Zhang

et al. 2020

IJMS

View full text Add to dashboard Cite

With recent advances in single-cell RNA sequencing, enormous transcriptome datasets have been generated. These datasets have furthered our understanding of cellular heterogeneity and its underlying mechanisms in homogeneous populations. Single-cell RNA sequencing (scRNA-seq) data clustering can group cells belonging to the same cell type based on patterns embedded in gene expression. However, scRNA-seq data are high-dimensional, noisy, and sparse, owing to the limitation of existing scRNA-seq technologies. Traditional clustering methods are not effective and efficient for high-dimensional and sparse matrix computations. Therefore, several dimension reduction methods have been introduced. To validate a reliable and standard research routine, we conducted a comprehensive review and evaluation of four classical dimension reduction methods and five clustering models. Four experiments were progressively performed on two large scRNA-seq datasets using 20 models. Results showed that the feature selection method contributed positively to high-dimensional and sparse scRNA-seq data. Moreover, feature-extraction methods were able to promote clustering performance, although this was not eternally immutable. Independent component analysis (ICA) performed well in those small compressed feature spaces, whereas principal component analysis was steadier than all the other feature-extraction methods. In addition, ICA was not ideal for fuzzy C-means clustering in scRNA-seq data analysis. K-means clustering was combined with feature-extraction methods to achieve good results.

show abstract

Adaptive community detection in complex networks using genetic algorithms

Cited by 111 publications

References 17 publications

The Central Community of Twitter ego-Networks as a Means for Fake Influencer Detection

The Central Community of Twitter ego-Networks as a Means for Fake Influencer Detection

The Analysis of Water Project Bid Rigging Behavior Based on Complex Network

Dimension Reduction and Clustering Models for Single-Cell RNA Sequencing Data: A Comparative Study

Contact Info

Product

Resources

About