TY - GEN
T1 - Combining one-versus-one and one-versus-all strategies to improve multiclass SVM classifier
AU - Chmielnicki, Wiesław
AU - Stąpor, Katarzyna
N1 - Publisher Copyright:
© Springer International Publishing Switzerland 2016.
PY - 2016
Y1 - 2016
N2 - Support Vector Machine (SVM) is a binary classifier, but most of the problems we find in the real-life applications are multiclass. There are many methods of decomposition such a task into the set of smaller classification problems involving two classes only. Two of the widely known are one-versus-one and one-versus-rest strategies. There are several papers dealing with these methods, improving and comparing them. In this paper, we try to combine theses strategies to exploit their strong aspects to achieve better performance. As the performance we understand both recognition ratio and the speed of the proposed algorithm. We used SVM classifier on several different databases to test our solution. The results show that we obtain better recognition ratio on all tested databases.Moreover, the proposed method turns out to be much more efficient than the original one-versus-one strategy.
AB - Support Vector Machine (SVM) is a binary classifier, but most of the problems we find in the real-life applications are multiclass. There are many methods of decomposition such a task into the set of smaller classification problems involving two classes only. Two of the widely known are one-versus-one and one-versus-rest strategies. There are several papers dealing with these methods, improving and comparing them. In this paper, we try to combine theses strategies to exploit their strong aspects to achieve better performance. As the performance we understand both recognition ratio and the speed of the proposed algorithm. We used SVM classifier on several different databases to test our solution. The results show that we obtain better recognition ratio on all tested databases.Moreover, the proposed method turns out to be much more efficient than the original one-versus-one strategy.
UR - https://www.scopus.com/pages/publications/84960906319
U2 - 10.1007/978-3-319-26227-7_4
DO - 10.1007/978-3-319-26227-7_4
M3 - Conference contribution
AN - SCOPUS:84960906319
SN - 9783319262253
T3 - Advances in Intelligent Systems and Computing
SP - 37
EP - 45
BT - Proceedings of the 9th International Conference on Computer Recognition Systems, CORES 2015
A2 - Burduk, Robert
A2 - Jackowski, Konrad
A2 - Kurzyński, Marek
A2 - Woźniak, Michał
A2 - Żołnierek, Andrzej
PB - Springer Verlag
T2 - 9th International Conference on Computer Recognition Systems, CORES 2015
Y2 - 25 May 2015 through 27 May 2015
ER -