2007年
A probabilistic decoding approach to multi-class classification
2007 IEEE INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS, VOLS 1-6
- ,
- 巻
- 号
- 1696
- 開始ページ
- 2670
- 終了ページ
- 2675
- 記述言語
- 英語
- 掲載種別
- DOI
- 10.1109/IJCNN.2007.4371380
- 出版者・発行元
- IEEE
In this article, we propose a new method of multiclass classification in the framework of error-correcting output coding (ECOC). Misclassification of each binary classifier is formulated as a bit inversion error with a probabilistic model for each class and dependence between binary classifiers is incorporated into our model, which makes a decoder, a type of Boltzmann machine. Experimental studies using a synthetic dataset and datasets from UCI repository are performed, and the results show that the proposed method is superior to other existing multi-class classification methods.
- リンク情報
- ID情報
-
- DOI : 10.1109/IJCNN.2007.4371380
- ISSN : 1098-7576
- Web of Science ID : WOS:000254291102101