Discovering overlapping communities in ego-nets using friend intimacy

Chang, Furong; Zhang, Bofeng; Li, Haiyan; Huang, Mingqing; Li, Bingchun; Zhao, Yue

doi:10.3233/jifs-172242

Cited by 5 publications

(7 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When the similarity reaches a certain level, there will be excessive overlapping, which results in the community redundancy, so it is necessary to further merge and optimize high similar communities. The previous methods (Chang et al, 2019; Jia et al, 2021; Liu et al, 2021; Liu & Wang, 2021; Xie et al, 2013) of community merging are to calculate the proportion of common nodes and use it to find the threshold of community merging, such as the merging method shown in Equation (). However, such methods ignore the influence of interest preference between different nodes in the process of community merging and the tightness of merged community.…”

Section: The Proposed Methods Eocmmmentioning

confidence: 99%

“…It can be seen from Figure 8 that our community merging method is better than the three merging method in

EQ

and

SEQ

, which proves that

SCOD

is effective and has certain advantages. In addition, in order to verify the overall effect of EOCMM in community detection, we compared three merging methods ENFI (Chang et al, 2019), NPLCE and GREESE on three datasets, and use the index

EQ

to evaluate the experimental results, as shown in Figure 9.…”

Section: Methodsmentioning

confidence: 99%

“…In this article, CPM (Palla et al, 2005), LFM (Lancichinetti et al, 2009), LFMJ (Yang et al, 2017), ENFI (Chang et al, 2019), NPLCE (Liu et al, 2021), GREESE (Asmi et al, 2022), WNOPM (Fu et al, 2021) and ASC (Berahmand et al, 2022) are selected as the comparison methods. Meanwhile, three evaluation indexes

EQ

D

and

SEQ

are utilized to evaluate the performance of each method.…”

Section: Methodsmentioning

confidence: 99%

“…Some scholars have studied the problem from the perspective of community redundancy. Lee et al (2010) judged whether to merge the community by the proportion of common nodes between the two communities, and part of the scholars (Chang et al, 2019; Li et al, 2020; Liu & Wang, 2021; Xie et al, 2013) have also used similar merging ideas. The local optimization overlapping community discovery algorithm combining attribute features method proposed by Jia et al (2021) defined a new fitness function by using the structural fitness between core nodes and noncore nodes, as well as the homogeneity and heterogeneity of communities.…”

Section: Related Workmentioning

confidence: 99%

“…In this article, CPM (Palla et al, 2005), LFM (Lancichinetti et al, 2009), LFMJ (Yang et al, 2017), ENFI (Chang et al, 2019), NPLCE , GREESE (Asmi et al, 2022), WNOPM (Fu et al, 2021) and ASC (Berahmand et al, 2022) are selected as the comparison methods.…”

Section: Overall Comparison Experimentsmentioning

confidence: 99%

See 4 more Smart Citations

An effective overlapping community merging method oriented to multidimensional attribute social networks

Cheng,

Yang,

Cheng

et al. 2023

Expert Systems

View full text Add to dashboard Cite

For multidimensional attribute social networks, most of the existing methods have limitations in dealing with community redundancy, namely, the influences of the network's multidimensional attributes and users' interest preferences are ignored in the process of community merging. To this end, we propose an effective overlapping community merging method (EOCMM) oriented to multidimensional attribute social networks. In EOCMM, we focus on two core problems which are how to improve the method of improving overlapping community detection in multidimensional attribute social networks and how to merge the detected communities with high redundancy degree. To solve the first problem, based on the network topology characteristics and user interest preference, the Node Domain index is proposed to improve and optimize the selection of seed nodes. The traditional overlapping community detection method is improved by semantic fitness function and seed nodes. To solve the second problem, we merge the communities with high redundancy degree by Semantic Community Overlapping Degree which is fused by user similarity, community similarity and community tightness. Finally, we compared our method with other multiple mainstream methods by extensive experiments on three real social network datasets. The experimental comparison results show that the improvements on three evaluation metrics extension of modularity, partition density and semantic extension of modularity are 6.04%, 7.96% and 3.13%, respectively. And our method can make the results of community detection and merging more reasonable and effective.

show abstract