抄録
This paper presents a new strategy to build multi tree hierarchical structure SVM which can get a more efficient and accuracy classification model for multiclass problems. Base on the theory of Binary Tree SVM (BTS), we proposed an improvement algorithm which extend binary tree structure to a multi tree structure, In the multi tree hierarchical structure, similarity clustering method was proposed to cluster classes to groups in each non-leaf node. In order to get a multi node division, one-against-all (OAA) was applied to train those groups rather than classes. The proposed method can avoid data imbalanced problem occurred in OAA, also the classification area of classifier in the upper layer is larger than classifier in lower layer. Compared with other several well-known methods, experiments on many data sets demonstrate that our method can reduce the number of classifiers in the testing phase and get a higher accuracy.
本文言語 | English |
---|---|
ホスト出版物のタイトル | Proceedings of the International Joint Conference on Neural Networks |
出版社 | Institute of Electrical and Electronics Engineers Inc. |
巻 | 2015-September |
ISBN(印刷版) | 9781479919604, 9781479919604, 9781479919604, 9781479919604 |
DOI | |
出版ステータス | Published - 2015 9 28 |
イベント | International Joint Conference on Neural Networks, IJCNN 2015 - Killarney, Ireland 継続期間: 2015 7 12 → 2015 7 17 |
Other
Other | International Joint Conference on Neural Networks, IJCNN 2015 |
---|---|
Country | Ireland |
City | Killarney |
Period | 15/7/12 → 15/7/17 |
ASJC Scopus subject areas
- Software
- Artificial Intelligence