In this paper we apply three pattern recognition methods (support vector machine, cluster analysis and principal component analysis) to distinguish regulatory regions from coding a...
Rene te Boekhorst, Irina I. Abnizova, Lorenz Werni...
In this paper, we extend the conventional vector quantization by incorporating a vigilance parameter, which steers the tradeoff between plasticity and stability during incremental...
Clustering with partial supervision finds its application in situations where data is neither entirely nor accurately labeled. This paper discusses a semisupervised clustering algo...
—A new formulation for multiway spectral clustering is proposed. This method corresponds to a weighted kernel principal component analysis (PCA) approach based on primal-dual lea...
A novel algorithm is proposed to learn pattern similarities for texture image retrieval. Similar patterns in di erent texture classes are grouped into a cluster in the feature spac...