Detecting intrinsic communities in evolving networks

Nath, Keshab; Roy, Swarup

doi:10.1007/s13278-019-0552-3

Cited by 13 publications

(4 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Table 11: Some weaknesses and strengths of incremental methods based on modularity optimization, density, and label propagation [52,30,40,63,4,46,18,10,70,16,55,6,68,47,54,42,36,8,57,65] 5.4 Advantages and disadvantages of incremental methods for community detection in both fully and growing dynamic networks…”

Section: Methods Based On Weaknesses Strengthsmentioning

confidence: 99%

“…These methods used also snapshots when identifying groups, taking into account the communities found in the previous snapshot but avoiding the need to match them. Other methods designed for detecting community in dynamic networks work directly on temporal networks are the incremental approaches [10,42,57]. They start with an initial community, and then update for each incoming change the community structure.…”

Section: Historic For Community Detectionmentioning

confidence: 99%

“…Table 8: Incremental community detection methods in growing dynamic networks based on density [10,42,8] Some abbreviations in Table 8 described in footnote: HOCTracker 25 , core node 26 Label propagation-based methods Hiroki et al [36] proposed a new algorithm to detect communities for high volume graphs stream based on DEMON 27 [19] algorithm. It used three functions for this issue, two incremental functions ((1) Ego minus ego network, (2) Label propagation) and (3) optimized merge function.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Incremental methods for community detection in both fully and growing dynamic networks

Bouhatem

Hadj

Souam

et al. 2021

Acta Universitatis Sapientiae, Informatica

View full text Add to dashboard Cite

In recent years, community detection in dynamic networks has received great interest. Due to its importance, many surveys have been suggested. In these surveys, the authors present and detail a number of methods that identify a community without taking into account the incremental methods which, in turn, also take an important place in dynamic community detection methods. In this survey, we provide a review of incremental approaches to community detection in both fully and growing dynamic networks. To do this, we have classified the methods according to the type of network. For each type of network, we describe three main approaches: the first one is based on modularity optimization; the second is based on density; finally, the third is based on label propagation. For each method, we list the studies available in the literature and state their drawbacks and advantages.

show abstract

Section: Methods Based On Weaknesses Strengthsmentioning

confidence: 99%

Section: Historic For Community Detectionmentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

Incremental methods for community detection in both fully and growing dynamic networks

Bouhatem

Hadj

Souam

et al. 2021

Acta Universitatis Sapientiae, Informatica

View full text Add to dashboard Cite

show abstract

“…Li et al (2020) are inspired to iteratively remove the reflection of a single image captured through a glass surface, in which they regard the transmission as the strong and dominant structure while the reflection the weak and hidden structure. Nath and Roy (2019) focus on the intrinsic community and extend the hidden community detection to dynamic networks. Gong et al (2018) propose multi-granularity community detection method (MGCD) based on network embedding to detect the hidden communities.…”

Section: Introductionmentioning

confidence: 99%

Structure Amplification on Multi-layer Stochastic Block Models

Xin,

He,

Bao

et al. 2021

Preprint

View full text Add to dashboard Cite

Much of the complexity of social, biological, and engineered systems arises from a network of complex interactions connecting many basic components. Network analysis tools have been successful at uncovering latent structure termed communities in such networks. However, some of the most interesting structure can be difficult to uncover because it is obscured by the more dominant structure. Our previous work proposes a general structure amplification technique called HICODE that uncovers many layers of functional hidden structure in complex networks. HICODE incrementally weakens dominant structure through randomization allowing the hidden functionality to emerge, and uncovers these hidden structure in real-world networks that previous methods rarely uncover. In this work, we conduct a comprehensive and systematic theoretical analysis on the hidden community structure. In what follows, we define multi-layer stochastic block model, and provide theoretical support using the model on why the existence of hidden structure will make the detection of dominant structure harder compared with equivalent random noise. We then provide theoretical proofs that the iterative reducing methods could help promote the uncovering of hidden structure as well as boosting the detection quality of dominant structure.

show abstract