Prediction of Chinese word-formation patterns using the layer-weighted semantic graph-based KFP-MCO classifier
详细信息 查看全文
- 作者：Guangxia Gaoa ; Zhiwang Zhangb ; zzwmis@163.com" class="auth_mail" title="E-mail the corresponding author
- 关键词：Semantic ; Word-formation ; Graph edit distance ; Multi-criteria optimization ; Pattern classification
- 刊名：Computer Speech & Language
- 出版年：2016
- 出版时间：September 2016
- 年：2016
- 卷：39
- 期：Complete
- 页码：29-46
- 全文大小：8039 K
文摘
- •
A novel layered semantic graph of Chinese semantic word-formation is proposed.
- •
The layer-weighted graph edit distance and the similarity kernel are defined.
- •
A new algorithm based on KFP-MCOC is used to predict word-formation patterns.
- •
Comparison of predictive performance is conducted between KFP-MCOC and SVM.
- •
Statistical test showed that the accuracy of the proposed approach is significant.