A task decomposition algorithm using mixtures of normal distributions for classification problems

Seiji Ishihara, Harukazu Igarashi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This paper proposes an algorithm for decomposing a multi-class classification problem into a set of two-class classification problems. The algorithm divides a set of input pattern vectors in each class into subsets according to the distribution of the selected input pattern vectors. The distribution is represented by a mixture of normal distributions, and the number of subsets is defined by using MDL criterion. The algorithm can be applied for constructing an effective modular neural network. We show also the experimental results of the construction and the advantages of the algorithm.

Original languageEnglish
Title of host publicationProceedings - Sixth International Conference on Hybrid Intelligent Systems and Fourth Conference on Neuro-Computing and Evolving Intelligence, HIS-NCEI 2006
DOIs
Publication statusPublished - 2006
Event6th International Conference on Hybrid Intelligent Systems and 4th Conference on Neuro-Computing and Evolving Intelligence, HIS-NCEI 2006 - Auckland
Duration: 2006 Dec 132006 Dec 15

Other

Other6th International Conference on Hybrid Intelligent Systems and 4th Conference on Neuro-Computing and Evolving Intelligence, HIS-NCEI 2006
CityAuckland
Period06/12/1306/12/15

    Fingerprint

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

Ishihara, S., & Igarashi, H. (2006). A task decomposition algorithm using mixtures of normal distributions for classification problems. In Proceedings - Sixth International Conference on Hybrid Intelligent Systems and Fourth Conference on Neuro-Computing and Evolving Intelligence, HIS-NCEI 2006 [4041408] https://doi.org/10.1109/HIS.2006.264911