Back

Quantitative evaluation of internal clustering validation indices using binary datesets

Pakgohar, N.; Lengyel, A.; Botta-Dukat, Z.

2023-08-12 ecology
10.1101/2023.08.09.552566 bioRxiv
Show abstract

Different clustering methods often classify the same dataset differently. Selecting the best clustering solution out of a multitude of alternatives is possible with cluster validation indices. The behavior of validity indices changes with the structure of the sample and the properties of the clustering algorithm. Unique properties of each index cause increasing or decreasing performance in some conditions. Due to the large variety of cluster validation indices, choosing the most suitable index concerning the dataset and clustering algorithms is challenging. We aim to assess different internal clustering validation indices. In the present paper, the validity indices consist of geometric and non-geometric methods. For this purpose, we applied simulated datasets with different noise levels. Each dataset was repeated 20 times. Three clustering algorithms with Jaccard dissimilarity are used, and 27 clustering validation indices are evaluated. The results provide a reliability guideline for the selection cluster validity indices.

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

1
Ecological Informatics
33 papers in training set
Top 0.1%
26.3%
2
Methods in Ecology and Evolution
176 papers in training set
Top 0.2%
12.6%
3
Ecological Indicators
21 papers in training set
Top 0.1%
11.8%
50% of probability mass above
4
PLOS ONE
5266 papers in training set
Top 23%
7.2%
5
PeerJ
308 papers in training set
Top 1%
4.8%
6
Ecology and Evolution
267 papers in training set
Top 2%
4.3%
7
Ecography
54 papers in training set
Top 0.5%
3.1%
8
Scientific Reports
3612 papers in training set
Top 40%
2.6%
9
Peer Community Journal
281 papers in training set
Top 2%
2.1%
10
Frontiers in Ecology and Evolution
69 papers in training set
Top 1%
1.7%
11
Ecosphere
57 papers in training set
Top 0.7%
1.7%
12
Oikos
84 papers in training set
Top 1.0%
1.5%
13
Environmental DNA
56 papers in training set
Top 0.4%
1.5%
14
Global Ecology and Biogeography
47 papers in training set
Top 0.7%
1.1%
15
Frontiers in Plant Science
256 papers in training set
Top 4%
1.1%
16
Ecological Modelling
28 papers in training set
Top 0.4%
1.0%
17
PLOS Computational Biology
1863 papers in training set
Top 18%
1.0%
18
Population Ecology
10 papers in training set
Top 0.2%
1.0%
19
Landscape Ecology
13 papers in training set
Top 0.3%
1.0%
20
Royal Society Open Science
214 papers in training set
Top 6%
0.8%
21
Journal of Biogeography
46 papers in training set
Top 0.8%
0.8%
22
Sustainability
10 papers in training set
Top 0.6%
0.6%
23
Frontiers in Microbiology
427 papers in training set
Top 9%
0.6%