Probabilistic classifiers are among the most popular classification methods adopted by the machine learning community. They are often based on a-priori knowledge about the probability distribution underlying the data; nevertheless this information is rarely provided, so that a family of probability distribution functions is assumed to be an approximation model. In this paper we present an efficient binary classifi cation algorithm, called Perceptron-IPCAC (PIPCAC), assuming that each class is distributed accordingly to a Mixture of Gaussian functions. PIPCAC is defined as a multilayer perceptron trained by combining different linear classifiers. The algorithm has been tested on both synthetic and real datasets, and the obtained results demonstrate the effectiveness and efficiency of the proposed method. Furthermore, the promising performances have been confirmed by the comparison of its results with those achieved by Support Vector Machines.
PIPCAC: A Novel Binary Classifier Assuming Mixtures of Gaussian Functions / A. Rozza, G. Lombardi, E. Casiraghi - In: Artificial Intelligence and Applications 2010 / [a cura di] M.H. Hamza. - Calgari, Canada : Acta Press, 2010. - ISBN 9780889868175. - pp. 104-111
PIPCAC: A Novel Binary Classifier Assuming Mixtures of Gaussian Functions
A. Rozza;G. Lombardi;E. Casiraghi
2010
Abstract
Probabilistic classifiers are among the most popular classification methods adopted by the machine learning community. They are often based on a-priori knowledge about the probability distribution underlying the data; nevertheless this information is rarely provided, so that a family of probability distribution functions is assumed to be an approximation model. In this paper we present an efficient binary classifi cation algorithm, called Perceptron-IPCAC (PIPCAC), assuming that each class is distributed accordingly to a Mixture of Gaussian functions. PIPCAC is defined as a multilayer perceptron trained by combining different linear classifiers. The algorithm has been tested on both synthetic and real datasets, and the obtained results demonstrate the effectiveness and efficiency of the proposed method. Furthermore, the promising performances have been confirmed by the comparison of its results with those achieved by Support Vector Machines.Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.