Consensus Clustering - Why Consensus Clustering?

Why Consensus Clustering?

• There are potential shortcomings for each of the known clustering techniques.
• Interpretation of results are difficult in a few cases.
• When there is no knowledge about the number of clusters, it becomes difficult.
• They are extremely sensitive to the initial settings.
• Some algorithms can never undo what was done previously.
• Iterative descent clustering methods, such as the SOM and K-means clustering circumvent some of the shortcomings of Hierarchical clustering by providing for univocally defined clusters and cluster boundaries. However, they lack the intuitive and visual appeal of Hierarchical clustering, and the number of clusters must be chosen a priori.
• An extremely important issue in cluster analysis is the validation of the clustering results, that is, how to gain confidence about the significance of the clusters provided by the clustering technique, (cluster numbers and cluster assignments). Lacking an external objective criterion (the equivalent of a known class label in supervised learning) this validation becomes somewhat elusive.