A4 Article in conference proceedings
Improving the Competency of Classifiers through Data Generation (2001)
Skrypnyk, I., & Viktor, H. (2001). Improving the Competency of Classifiers through Data Generation. In V. Kůrková, R. Neruda, M. Kárný, & N. C. Steele (Eds.), Proceedings 5th Int. Conf. on Artificial neural Networks and Genetic Algorithms ICANNGA'01, Prague, Czech Republic, April 2001. Springer. https://doi.org/10.1007/978-3-7091-6230-9_13
JYU authors or editors
Publication details
All authors or editors: Skrypnyk, Iryna; Viktor, Herna
Parent publication: Proceedings 5th Int. Conf. on Artificial neural Networks and Genetic Algorithms ICANNGA'01, Prague, Czech Republic, April 2001
Parent publication editors: Kůrková, Věra; Neruda, Roman; Kárný, Miroslav; Steele, Nigel C.
Place and date of conference: Prague, Czech Republic, 2001
ISBN: 978-3-211-83651-4
eISBN: 978-3-7091-6230-9
Publication year: 2001
Number of pages in the book: 506
Publisher: Springer
Place of Publication: Vienna
Publication country: Austria
Publication language: English
DOI: https://doi.org/10.1007/978-3-7091-6230-9_13
Publication open access: Not open
Publication channel open access:
Abstract
This paper describes a hybrid approach in which sub-symbolic neural networks and symbolic machine learning algorithms are grouped into an ensemble of classifiers. Initially each classifier determines which portion of the data it is most competent in. The competency information is used to generated new data that are used for further training and prediction. The application of this approach in a difficult to learn domain shows an increase in the predictive power, in terms of the accuracy and level of competency of both the ensemble and the component classifiers.
Free keywords: problem domain; training instance; data generation process; disjunctive normal form; decision tree algorithm
Contributing organizations
Ministry reporting: Yes
Preliminary JUFO rating: Not rated