In this abstract, we cope with the problem of Model Selection (MS) of Support Vector Machine (SVM) classiers in the Online Learning (OL) framework. Though often neglected in OL, MS is paramount to guarantee that model eectiveness is guaranteed while exploiting new gathered samples. In particular we propose a heuristic approach, which can be feasibly applied in OL applications to SVM. The eectiveness of the proposal is supported by preliminary experimental results.
A Heuristic Approach to Model Selection for Online Support Vector Machines
ANGUITA, DAVIDE;GHIO, ALESSANDRO;LAWAL, ISAH ABDULLAHI;ONETO, LUCA
2013-01-01
Abstract
In this abstract, we cope with the problem of Model Selection (MS) of Support Vector Machine (SVM) classiers in the Online Learning (OL) framework. Though often neglected in OL, MS is paramount to guarantee that model eectiveness is guaranteed while exploiting new gathered samples. In particular we propose a heuristic approach, which can be feasibly applied in OL applications to SVM. The eectiveness of the proposal is supported by preliminary experimental results.
File in questo prodotto:
Non ci sono file associati a questo prodotto.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
