A Validity Index for Fuzzy Clustering Based on Bipartite Modularity

Table 2

The detailed information of datasets.

Datasets	Samples	Attributes	Clusters	Description
Iris	150	4	3	The dataset is the most famous database in the pattern recognition literatures, including 150 samples, 4 attributes, and 3 classes containing 50 instances each, and each class refers to a type of iris.
Wine	178	13	3	The dataset is the result of chemical analysis on three different kinds of wines from the same region of Italy. It contains 178 results. The analysis determines the number of 13 components found in each wine. The dataset is divided into three categories.
Wpbc	194	33	2	The dataset is a Wisconsin breast cancer dataset with 194 samples and 33 attributes, and each sample represents a subsequent data of breast cancer case. The dataset is divided into two categories.
Hayesroth	132	5	3	The dataset describes the evidence of behaviour classification, confidence degree, and recognition behaviour that are closely related to human behaviour. It contains 132 samples and 5 attributes using numerical values instead of actual values. The dataset is divided into three classes.
Zoo	101	16	7	The dataset describes the classification of animals, including 101 samples and 17 attributes. The dataset is divided into 7 classes.
Glass	214	9	6	The study of glass classification is motivated by criminological investigations. If the identification is correct, the remaining glass can act as evidence. The dataset consists of 214 samples and 9 attributes, which is divided into 6 classes.