An Empirical Comparison of Algorithms to Find Communities in Directed Graphs and Their Application in Web Data Analytics

Agreste, Santa; Meo, Pasquale De; Fiumara, Giacomo; Piccione, Giuseppe; Piccolo, Sebastiano; Rosaci, Domenico; Sarné, Giuseppe M. L.; Vasilakos, Athanasios V.

doi:10.1109/tbdata.2016.2631512

Cited by 22 publications

(16 citation statements)

References 83 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In terms of speed, label propagation [20] is capable of detecting communities in large-scale networks near linear time, though the solutions are usually nonunique. Additionally, other approaches such as Walk-Trap [21], Infomap [22], Louvain [23], and their empirical competitiveness are subjected to trade-off between accuracy and scalability [24]. Representation learning methods such as GraRep [25] and CFOND [26] consider the completion of their adjacency matrix and can be generally considered as matrix factorization problem.…”

Section: Related Workmentioning

confidence: 99%

Variational Approach for Learning Community Structures

2018

View full text Add to dashboard Cite

Discovering and modeling community structure exist to be a fundamentally challenging task. In domains such as biology, chemistry, and physics, researchers often rely on community detection algorithms to uncover community structures from complex systems yet no unified definition of community structure exists. Furthermore, existing models tend to be oversimplified leading to a neglect of richer information such as nodal features. Coupled with the surge of user generated information on social networks, a demand for newer techniques beyond traditional approaches is inevitable. Deep learning techniques such as network representation learning have shown tremendous promise. More specifically, supervised and semisupervised learning tasks such as link prediction and node classification have achieved remarkable results. However, unsupervised learning tasks such as community detection remain widely unexplored. In this paper, a novel deep generative model for community detection is proposed. Extensive experiments show that the proposed model, empowered with Bayesian deep learning, can provide insights in terms of uncertainty and exploit nonlinearities which result in better performance in comparison to state-of-the-art community detection methods. Additionally, unlike traditional methods, the proposed model is community structure definition agnostic. Leveraging on low-dimensional embeddings of both network topology and feature similarity, it automatically learns the best model configuration for describing similarities in a community.

show abstract

Section: Related Workmentioning

confidence: 99%

Variational Approach for Learning Community Structures

2018

View full text Add to dashboard Cite

show abstract

“…However, the real-world complex network often resembles the directed graph. Agreste et al have made an extensive comparison of community detection algorithms for the directed network [27]. Infomap and Label propagation algorithms are the first ones to implement in the directed network.…”

Section: Related Workmentioning

confidence: 99%

Large Scale Community Detection Using a Small World Model

et al. 2017

View full text Add to dashboard Cite

Abstract:In a social network, small or large communities within the network play a major role in deciding the functionalities of the network. Despite of diverse definitions, communities in the network may be defined as the group of nodes that are more densely connected as compared to nodes outside the group. Revealing such hidden communities is one of the challenging research problems. A real world social network follows small world phenomena, which indicates that any two social entities can be reachable in a small number of steps. In this paper, nodes are mapped into communities based on the random walk in the network. However, uncovering communities in large-scale networks is a challenging task due to its unprecedented growth in the size of social networks. A good number of community detection algorithms based on random walk exist in literature. In addition, when large-scale social networks are being considered, these algorithms are observed to take considerably longer time. In this work, with an objective to improve the efficiency of algorithms, parallel programming framework like Map-Reduce has been considered for uncovering the hidden communities in social network. The proposed approach has been compared with some standard existing community detection algorithms for both synthetic and real-world datasets in order to examine its performance, and it is observed that the proposed algorithm is more efficient than the existing ones.

show abstract

“…Malliaros and Vazirgiannis classify several approaches used to graph clustering and community detection on the directed network: (i) naive graph transformation approaches, (ii) transformations maintaining directionality, (iii) approaches that extend objective functions and methodologies to directed networks, and (iv) alternative approaches [18]. Some algorithms used to detect communities on directed graphs are the propagation label algorithm and the infomap algorithm [1]. Tests performed by [1] artificial and real graph data, propagation label algorithms have good performance in scalability, while the infomap algorithm has the best performance for accuracy and computational performance [1].…”

Section: Introductionmentioning

confidence: 99%

“…Some algorithms used to detect communities on directed graphs are the propagation label algorithm and the infomap algorithm [1]. Tests performed by [1] artificial and real graph data, propagation label algorithms have good performance in scalability, while the infomap algorithm has the best performance for accuracy and computational performance [1].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Review on Overlapping and Non-Overlapping Community Detection Algorithms for Social Network Analytics

negara¹,

Andriani²

2018

FJEC

View full text Add to dashboard Cite

Community detection is a common problem that exists in the data graph analytics as the social networks analytics. In the context of social networks, community detection is aimed at to find a group or community that has connectedness between individuals (nodes) with high-intensity interaction. In general, types of group on the social networking can be divided into the overlapping and non-overlapping. We provide an overview of some of the algorithms of overlapping and non-overlapping community detections available today to perform an analysis or a breakdown of the data of social networking. The algorithms for overlapping community detection are: (1) local seed selection algorithm; (2) seed set expansion algorithm; (3) speaker listener label propagation algorithm (SPLA) and the algorithms for detection of non-overlapping community are: (4) multithreaded Edi Surya Negara and Ria Andryani 2 community detection algorithm (MCD); (5) core groups graph cluster algorithm (CGGC); (6) complex network cluster detection (CONCLUDE); (7) dense subgraph extraction (DSE).

show abstract

An Empirical Comparison of Algorithms to Find Communities in Directed Graphs and Their Application in Web Data Analytics

Cited by 22 publications

References 83 publications

Variational Approach for Learning Community Structures

Variational Approach for Learning Community Structures

Large Scale Community Detection Using a Small World Model

A Review on Overlapping and Non-Overlapping Community Detection Algorithms for Social Network Analytics

Contact Info

Product

Resources

About