Evolutionary Optimization of Meta Data Metric for Method Recommendation

Autoři: Mgr. Roman Neruda, Ph.D.
Rok vydání: 2013
Typ publikace: Konferenční příspěvek (mezinárodní konf.)
Periodikum: Proceedings of the 2013 IEEE Conference on Cybernetics and Intelligent Systems (CIS)
Projekty: Učení funkcionálních vztahů z vysoce dimenzionálních dat
Bibliografická citace: KAZÍK, O., ŠMÍD, J., NERUDA, Roman. Evolutionary Optimization of Meta Data Metric for Method Recommendation. In: Proceedings of the 2013 IEEE Conference on Cybernetics and Intelligent Systems (CIS). Piscataway: IEEE Computer Society, 2013, S. 123-127. ISBN 978-1-4799-1072-4
Dostupné z: doi:10.1109/ICCIS.2013.6751590
DOI: 10.1109/ICCIS.2013.6751590
ID: 425750

Abstrakt

Metalearning - a method for recommendation the most suitable data-mining algorithm to an unknown dataset - is an important problem that needs to be solved in order to design a completely autonomous data-mining solver. This paper deals with this particular problem by proposing a machinelearning method which recommends the most suitable algorithm to an unknown dataset based on the results of previous datamining experiments. The fundamental idea behind this is that the algorithms will perform similarly on similar datasets. The choice of datasets features - called meta data - is presented and the metric comparing datasets is optimized by means of evolutionary computation.