Abstract. Discriminative and generative methods provide two distinct approaches to machine learning classification. One advantage of generative approaches is that they naturally mo...
Visualization techniques provide an outstanding role in KDD process for data analysis and mining. However, one image does not always convey successfully the inherent information fr...
Classification is one of the basic tasks of data mining in modern database applications including molecular biology, astronomy, mechanical engineering, medical imaging or meteorolo...
The rigid all-versus-all adjacency required by a maximal biclique for its two vertex sets is extremely vulnerable to missing data. In the past, several types of quasi-bicliques ha...
Manifold learning is an effective methodology for extracting nonlinear structures from high-dimensional data with many applications in image analysis, computer vision, text data a...