Unsupervized data-driven partitioning of multiclass problems

H. AHUMADA; G.L. GRINBLAT; P. M. GRANITTO

Espoo

Congreso; ICANN'11 International Conference on Artificial Neural Networks (LNCS); 2011

ENNS

Many classification problems of high technological value aremulticlass. In the last years, several improved solutions based on thecombination of simple classifiers were introduced. An interesting kind of methods creates a hierarchy of sub-problems by clustering prototypes of each one of the classes, but the solution produced by the clustering stage is heavily influenced by the label’s information. In this work we introduce a new strategy to solve multiclass problems that makes more use of spatial information than other methods. Based on our previous work on imbalanced problems, we construct a hierarchy of subproblems, but opposite to previous developments, based only on spatial information and not using class labels at any time. We consider different clustering methods (either agglomerative or divisive) for this task. We use an SVM for each sub-problem (if needed, because in several cases the clustering method directly gives a subset with samples of a single class). Using publicly available datasets we compare the new method with several previous approaches, finding promising results.