Clustering Based Classification in Data Mining Method Recommendation

0425703 - ÚI 2015 RIV US eng C - Konferenční příspěvek (zahraniční konf.)
Kazík, O. - Pešková, K. - Šmíd, J. - Neruda, Roman
Clustering Based Classification in Data Mining Method Recommendation.
ICMLA 2013. Proceedings of the 12th International Conference on Machine Learning and Applications. Vol. 2. Los Alamitos: IEEE Computer Society, 2013 - (Wani, M.; Tecuci, G.; Boicu, M.; Kubát, M.; Khoshgoftaar, T.; Seliya, N.), s. 356-361. ISBN 978-0-7695-5144-9.
[ICMLA 2013. International Conference on Machine Learning and Applications /12./. Miami (US), 04.12.2013-07.12.2013]
Grant CEP: GA ČR GAP202/11/1368; GA MŠMT(CZ) LD13002
Grant ostatní: GA UK(CZ) 29612; SVV(CZ) 265314
Institucionální podpora: RVO:67985807
Klíčová slova: metalearning * clustering * data mining * method recommendation
Kód oboru RIV: IN - Informatika

With the growing amount of data available in today’s world, the emphasis is laid on the automatic configuration of data analysis – metalearning. This paper elaborates one of the metalearning subproblems, the data mining method recommendation. Based on a metric over the data features called metadata, we have proposed a solution exploiting clustering of datasets. The agglomerative algorithm is used to construct clustering over the metadata, and the average methods’ performance is computed in each cluster. The ranking of data mining methods is then deduced from the classification of a dataset to a particular cluster. The recommendation algorithm, which is implemented within our data mining multi-agent system, has been tested in various configurations, and the results of these experiments have been compared.
Trvalý link: http://hdl.handle.net/11104/0232431