Comparative evaluation of community detection algorithms: a topological approach (original) (raw)

Qualitative Comparison of Community Detection Algorithms

2011

Community detection is a very active field in complex networks analysis, consisting in identifying groups of nodes more densely interconnected relatively to the rest of the network. The existing algorithms are usually tested and compared on real-world and artificial networks, their performance being assessed through some partition similarity measure. However, artificial networks realism can be questioned, and the appropriateness of those measures is not obvious. In this study, we take advantage of recent advances concerning the characterization of community structures to tackle these questions. We first generate networks thanks to the most realistic model available to date. Their analysis reveals they display only some of the properties observed in real-world community structures. We then apply five community detection algorithms on these networks and find out the performance assessed quantitatively does not necessarily agree with a qualitative analysis of the identified communities. It therefore seems both approaches should be applied to perform a relevant comparison of the algorithms.

Analysis of Communities Detection Algorithms in Complex Networks

International Journal of Computer Applications

Complex networks are an imminent multidisciplinary field defined by graphs that present a nontrivial topographic structure. An important information extracted from a complex network is its communities structure. In the literature, there are several communities detection algorithms, however, new research have emerged with the aim of detecting communities efficiently and with lower computational cost. Therefore, this work analyzes different algorithms for communities detection in complex networks with different characteristics, considering the Modularity measure, the execution time and the obtained communities number. The partitions obtained by the different algorithms presented high modularity values and it was observed that the influence of the number of vertices and edges in the execution time of some detection algorithms.

A Comparison of Community Detection Algorithms on Artificial Networks

2009

Community detection has become a very important part in complex networks analysis. Authors traditionally test their algorithms on a few real or artificial networks. Testing on real networks is necessary, but also limited: the considered real networks are usually small, the actual underlying communities are generally not defined objectively, and it is not possible to control their properties. Generating artificial networks makes it possible to overcome these limitations. Until recently though, most works used variations of the classic Erdős-Rényi random model and consequently suffered from the same flaws, generating networks not realistic enough. In this work, we use Lancichinetti et al. model, which is able to generate networks with controlled power-law degree and community distributions, to test some community detection algorithms. We analyze the properties of the generated networks and use the normalized mutual information measure to assess the quality of the results and compare the considered algorithms.

Community detection algorithms: A comparative analysis

Physical Review E, 2009

Uncovering the community structure exhibited by real networks is a crucial step towards an understanding of complex systems that goes beyond the local organization of their constituents. Many algorithms have been proposed so far, but none of them has been subjected to strict tests to evaluate their performance. Most of the sporadic tests performed so far involved small networks with known community structure and/or artificial graphs with a simplified structure, which is very uncommon in real systems. Here we test several methods against a recently introduced class of benchmark graphs, with heterogeneous distributions of degree and community size. The methods are also tested against the benchmark by Girvan and Newman and on random graphs. As a result of our analysis, three recent algorithms introduced by Rosvall and Bergstrom, Blondel et al. and Ronhovde and Nussinov, respectively, have an excellent performance, with the additional advantage of low computational complexity, which enables one to analyze large systems.

Various Approaches of Community Detection in Complex Networks: A Glance

International Journal of Information Technology and Computer Science, 2016

Identifying strongly associated clusters in large complex networks has received an increased amount of interest since the past decade. The problem of community detection in complex networks is an NP complete problem that necessitates the clustering of a network into communities of compactly linked nodes in such a manner that the interconnection between the nodes is found to be denser than the intra-connection between the communities. In this paper, different approaches given by the authors in the field of community detection have been described with each methodology being classified according to algorithm type, along with the comparative analysis of these approaches on the basis of NMI and Modularity for four real world networks.

Community structures evaluation in complex networks: A descriptive approach

Evaluating a network partition just only via conventional quality metrics – such as modularity, conductance or normalized mutual of information – is usually insufficient. Indeed, global quality scores of a network partition or its clusters do not provide many ideas about their structural characteristics. Furthermore, quality metrics often fail to reach an agreement especially in networks whose modular structures are not very obvious. Evaluating the goodness of network partitions in function of desired structural properties is still a challenge. Here, we propose a methodology that allows one to expose structural information of clusters in a network partition in a comprehensive way, thus eventually helps one to compare communities identified by different community detection methods. This descriptive approach also helps to clarify the composition of communities in real-world networks. The methodology hence bring us a step closer to the understanding of modular structures in complex networks.

"Communities in Networks: An empirical and comparative Study of some Detection Methods"

Graphs or networks are mathematical structures that consist of elements that can be pairwise linked if some sort of interaction exists between them. Therefore, they can be used as an abstraction to represent complex systems in various fields of research. Several statistical properties are common to many real-world networks, either being natural or man-made. A specific feature also shared by many networks is that they often exhibit internal substructures, called communities, consisting of cohesive groups of elements which are densely connected to each other but only sparsely or not connected at all to other communities of the graph. Community detection is today an active and interdisciplinary field of research. This thesis focuses on criteria that could allow a reliable detection of these substructures and on methods based on these criteria. In the first research question, we will cover and compare the performance of methods that return an optimal number of communities: The so-called ... Document type : Mémoire (Thesis) Référence bibliographique Jungers, Baptiste. Communities in Networks: An empirical and comparative Study of some Detection Methods. Louvain School of Management, Université catholique de Louvain, 2016.

On Accuracy of Community Structure Discovery Algorithms

2011

Community structure discovery in complex networks is a quite challenging problem spanning many applications in various disciplines such as biology, social network and physics. Emerging from various approaches numerous algorithms have been proposed to tackle this problem. Nevertheless little attention has been devoted to compare their efficiency on realistic simulated data. To better understand their relative performances, we evaluate systematically eleven algorithms covering the main approaches. The Normalized Mutual Information (NMI) measure is used to assess the quality of the discovered community structure from controlled artificial networks with realistic topological properties. Results show that along with the network size, the average proportion of intra-community to inter-community links is the most influential parameter on performances. Overall, "Infomap" is the leading algorithm, followed by "Walktrap", "SpinGlass" and "Louvain" which also achieve good consistency.

Community detection algorithm evaluation with ground-truth data

Physica A: Statistical Mechanics and its Applications, 2018

Community structure is of paramount importance for the understanding of complex networks. Consequently, there is a tremendous effort in order to develop efficient community detection algorithms. Unfortunately, the issue of a fair assessment of these algorithms is a thriving open question. If the ground-truth community structure is available, various clustering-based metrics are used in order to compare it versus the one discovered by these algorithms. However, these metrics defined at the node level are fairly insensitive to the variation of the overall community structure. To overcome these limitations, we propose to exploit the topological features of the 'community graphs' (where the nodes are the communities and the links represent their interactions) in order to evaluate the algorithms. To illustrate our methodology, we conduct a comprehensive analysis of overlapping community detection algorithms using a set of real-world networks with known a priori community structure. Results provide a better perception of their relative performance as compared to classical metrics. Moreover, they show that more emphasis should be put on the topology of the community structure. We also investigate the relationship between the topological properties of the community structure and the alternative evaluation measures (quality metrics and clustering metrics). It appears clearly that they present different views of the community structure and that they must be combined in order to evaluate the effectiveness of community detection algorithms.