On Comparing and Enhancing Two Common Approaches to Network Community Detection

@article{Motschnig2021OnCA,
  title={On Comparing and Enhancing Two Common Approaches to Network Community Detection},
  author={Niko Motschnig and Alexander Ramharter and Oliver Schweiger and Philipp Zabka and Klaus-Tycho Foerster},
  journal={2021 IEEE Global Communications Conference (GLOBECOM)},
  year={2021},
  pages={1-6}
}
In this work, we explore two common algorithms for community detection in networks, namely Agglomerative Hierarchical Clustering and the Louvain Method. We investigate their mechanics and compare their differences in terms of implementation and results of the clustering behavior on a standard dataset. We further propose some enhancements to these algorithms that show promising results in our evaluations, such as self-neighboring for Neighbor Matrix constructions and a deterministic and slightly… 

Figures and Tables from this paper

A Graph Representation Learning Algorithm for Approximate Local Symmetry Feature Extraction to Enhance Malicious Device Detection Preprocessing

The Graph Structural-topic Similar Subgraph Merging, abbreviated GraphSTSGM, is proposed to extract topological similarity between nodes to enhance the ability of malicious device detection by accurately characterizing the approximate local symmetry features of nodes.

Oppression by omission: An analysis of the #WhereIsTheInterpreter hashtag campaign around COVID-19 on Twitter

Critical to managing a crisis such as COVID-19 is the propagation of information to all vulnerable populations. Despite guidelines regarding communicating with people with differing accessibility

References

SHOWING 1-10 OF 22 REFERENCES

Generalized Louvain method for community detection in large networks

A novel strategy to discover the community structure of (possibly, large) networks by exploiting a novel measure of edge centrality, based on the κ-paths, which allows to efficiently compute a edge ranking in large networks in near linear time.

Soft Clustering on Graphs

A hierarchical clustering is naturally derived in this framework to gradually merge lower-level clusters into higher-level ones in a probabilistic way to discover a more global clustering structure.

Hierarchical Graph Clustering using Node Pair Sampling

It is proved that this distance between clusters induced by the probability of sampling node pairs is reducible, which enables the use of the nearest-neighbor chain to speed up the agglomeration.

A Greedy Algorithm for Neighborhood Overlap-Based Community Detection

This paper proposes a greedy algorithm of iteratively removing the edges of a network in the increasing order of their neighborhood overlap and calculating the modularity score of the resulting network component(s) after the removal of each edge.

Fast unfolding of communities in large networks

This work proposes a heuristic method that is shown to outperform all other known community detection methods in terms of computation time and the quality of the communities detected is very good, as measured by the so-called modularity.

A Comparative Analysis of Community Detection Algorithms on Artificial Networks

This study uses the Lancichinetti-Fortunato-Radicchi benchmark graph to test eight state-of-the-art community detection algorithms and provides guidelines that help to choose the most adequate community detection algorithm for a given network.

Community Detection : Hierarchical clustering Algorithms

Hierarchical clustering algorithms and software’s implementing them for community detection are discussed in this paper and data mining, community detection, hierarchical clustering is discussed.

Fast algorithm for detecting community structure in networks.

  • M. Newman
  • Computer Science
    Physical review. E, Statistical, nonlinear, and soft matter physics
  • 2004
An algorithm is described which gives excellent results when tested on both computer-generated and real-world networks and is much faster, typically thousands of times faster, than previous algorithms.

Computing Communities in Large Networks Using Random Walks

A measure of similarities between vertices based on random walks which captures well the community structure in a network, can be computed efficiently, and it can be used in an agglomerative algorithm to compute efficiently thecommunity structure of a network.

Defining and evaluating network communities based on ground-truth

It is argued that the goal of network community detection is to extract functional communities based on the connectivity structure of the nodes in the network, and a methodology is proposed, which allows for compare and quantitatively evaluate how different structural definitions of communities correspond to ground-truth functional communities.