Classification of experimental data by simple and composed classifiers

An important part of decision tasks is classification of objects into classes. If there is a set of input data, which class memberships are known, based on these data it is possible to take a decision on membership of new data of the same type. Nowadays many classification technologies and algorithms are developed. Increased requirements are taken on these technologies in regard to increased precision, shorter classification time and so on. This contribution deals with simple - k-nearest neighbours, Bayesian classifier, decision tree and composed classifiers - Bagging, Boosting and Stacked Generalization applied on experimental data set.