Unsupervised Optimal Fuzzy Clustering
IEEE Transactions on Pattern Analysis and Machine Intelligence
A Validity Measure for Fuzzy Clustering
IEEE Transactions on Pattern Analysis and Machine Intelligence
Validating fuzzy partitions obtained through c-shells clustering
Pattern Recognition Letters - Special issue on fuzzy set technology in pattern recognition
A unified framework for model-based clustering
The Journal of Machine Learning Research
Fuzzy clustering of categorical data using fuzzy centroids
Pattern Recognition Letters
A cluster validation index for GK cluster analysis based on relative degree of sharing
Information Sciences—Informatics and Computer Science: An International Journal
A cluster validity index for fuzzy clustering
Pattern Recognition Letters
On fuzzy cluster validity indices
Fuzzy Sets and Systems
Fuzzy clustering for symbolic data
IEEE Transactions on Fuzzy Systems
A fuzzy k-modes algorithm for clustering categorical data
IEEE Transactions on Fuzzy Systems
Low-complexity fuzzy relational clustering algorithms for Web mining
IEEE Transactions on Fuzzy Systems
| Hi-index | 0.00 |
Clustering is an unsupervised learning method that partitions the objects in a given object set into clusters in which objects are similar. Iterative clustering algorithms have been widely applied in a variety of key areas. Those algorithms find clusters of a fixed given number. The number of clusters must be decided before the algorithms run. The number of clusters is usually obtained by using cluster validity index algorithms. There have been many studies for cluster validity index, especially for fuzzy clustering. However, almost all of the studies focus only on the clustering of the objects of numerical features, even though most of the real objects include ordinal and categorical features as well. In this paper, we identify which fuzzy cluster validity indices can be applied for the objects of mixed features. We experiment with these selected indices and one new index on various synthetic object sets and real object sets, in which objects have mixed features. We present a few indices out of many indices as the experiment results, which work better for the most of experiment cases.