Skip to Main Content
Presents several cluster evaluation techniques for gene expression data analysis. Normalisation and validity aggregation strategies are proposed to improve the prediction of the number of relevant clusters. The effect of different intracluster and intercluster distances on this prediction process is studied. This approach is applied to a publicly released medulloblastomas tumour data set The results suggest that it may represent an effective tool to support biomedical knowledge discovery tasks based on gene expression data.