Silhouettes: a graphical aid to the interpretation and validation of cluster analysis
Journal of Computational and Applied Mathematics
Combining Multiple Clusterings Using Evidence Accumulation
IEEE Transactions on Pattern Analysis and Machine Intelligence
Improved graph-based metrics for clustering high-dimensional datasets
IBERAMIA'10 Proceedings of the 12th Ibero-American conference on Advances in artificial intelligence
Hi-index | 0.00 |
In this work we introduce a modification to an automatic non-supervised rule to select the parameters of a previously presented graph-based metric. This rule maximizes a clustering quality index providing the best possible solution from a clustering quality point of view. We apply our parameter-free PKNNG metric on gene expression data to show that the best quality solutions are also the ones that are more related to the biological classes. Finally, we compare our parameter-free metric with a group of state-of-the-art clustering algorithms. Our results indicate that our parameter-free metric performs as well as the state-of-the-art clustering methods.