Abstract
Modularity Evaluation (ME) is usually used in community detection for evaluating the disjoint and overlapping communities. In this paper, two obvious defects of ME are revealed and proved, including the non-decreasing contribution of isolated nodes to modularity and lacking of appropriate measures on overlapping community. We also propose a new evaluation criterion, the USI-AUC, which is the Area Under the Curve (AUC), originated from link-prediction of Uniform-Structure-Information (USI) model. We test the new criterion on various datasets, and find that such criterion can avoid the issues exposed in ME.
| Original language | English |
|---|---|
| Pages (from-to) | 439-462 |
| Number of pages | 24 |
| Journal | Intelligent Data Analysis |
| Volume | 22 |
| Issue number | 2 |
| DOIs | |
| Publication status | Published - 14 Mar 2018 |
Fingerprint
Keywords
- Evaluation of communities
- link prediction
- modularity evaluation
- USI model
- USI-AUC criterion
Cite this
}
USI-AUC : an evaluation criterion of community detection based on a novel link-prediction method. / Wu, Yiteng; Yu, Hongtao; Zhang, Jianpeng; Liu, Shuxin; Huang, Ruiyang; Li, Peng.
In: Intelligent Data Analysis, Vol. 22, No. 2, 14.03.2018, p. 439-462.Research output: Contribution to journal › Article › Academic › peer-review
TY - JOUR
T1 - USI-AUC : an evaluation criterion of community detection based on a novel link-prediction method
AU - Wu, Yiteng
AU - Yu, Hongtao
AU - Zhang, Jianpeng
AU - Liu, Shuxin
AU - Huang, Ruiyang
AU - Li, Peng
PY - 2018/3/14
Y1 - 2018/3/14
N2 - Modularity Evaluation (ME) is usually used in community detection for evaluating the disjoint and overlapping communities. In this paper, two obvious defects of ME are revealed and proved, including the non-decreasing contribution of isolated nodes to modularity and lacking of appropriate measures on overlapping community. We also propose a new evaluation criterion, the USI-AUC, which is the Area Under the Curve (AUC), originated from link-prediction of Uniform-Structure-Information (USI) model. We test the new criterion on various datasets, and find that such criterion can avoid the issues exposed in ME.
AB - Modularity Evaluation (ME) is usually used in community detection for evaluating the disjoint and overlapping communities. In this paper, two obvious defects of ME are revealed and proved, including the non-decreasing contribution of isolated nodes to modularity and lacking of appropriate measures on overlapping community. We also propose a new evaluation criterion, the USI-AUC, which is the Area Under the Curve (AUC), originated from link-prediction of Uniform-Structure-Information (USI) model. We test the new criterion on various datasets, and find that such criterion can avoid the issues exposed in ME.
KW - Evaluation of communities
KW - link prediction
KW - modularity evaluation
KW - USI model
KW - USI-AUC criterion
UR - http://www.scopus.com/inward/record.url?scp=85044389980&partnerID=8YFLogxK
U2 - 10.3233/IDA-173400
DO - 10.3233/IDA-173400
M3 - Article
AN - SCOPUS:85044389980
VL - 22
SP - 439
EP - 462
JO - Intelligent Data Analysis
JF - Intelligent Data Analysis
SN - 1088-467X
IS - 2
ER -