Co-evolutionary Genetic Programming for Dataset Similarity Induction

0459144 - ÚI 2017 RIV US eng C - Konferenční příspěvek (zahraniční konf.)
Šmíd, J. - Pilát, M. - Pešková, K. - Neruda, Roman
Co-evolutionary Genetic Programming for Dataset Similarity Induction.
2015 IEEE Congress on Evolutionary Computation. Piscataway: IEEE CS, 2015, s. 1160-1166. ISBN 978-1-4799-7492-4. ISSN 1089-778X. E-ISSN 1941-0026.
[CEC 2015. IEEE Congress on Evolutionary Computation. Sendai (JP), 25.05.2015-28.05.2015]
Grant CEP: GA ČR GA15-19877S
Institucionální podpora: RVO:67985807
Klíčová slova: metalearning * genetic programming * datamining * co-evolution * metric
Kód oboru RIV: IN - Informatika

Metalearning deals with an important problem in machine-learning, namely selecting the right techniques to model the data at hand. In most of the metalearning approaches, a notion of similarity between datasets is needed. Our approach derives the similarity measure by combining arbitrary attribute similarity functions ordered by the optimal attribute assignment. In this paper, we propose a genetic programming based approach to the evolution of an attribute similarity inducing function. The function is composed of two parts - one describes the similarity of categorical attributes, the other describes the similarity of numerical attributes. Co-evolution is used to put these two parts together to form the similarity function. We use a repairing approach to guarantee some of the metric features for this function, and also discuss which of these features are important in metalearning.
Trvalý link: http://hdl.handle.net/11104/0259384