Neural network model for dependency parsing incorporating global vector feature
LSTM and piecewise CNN were utilized to extract word vector features and global vector features,respectively.Then the two features were input to feed forward network for training.In model training,the probabilistic training method was adopted.Compared with the original dependency paring model,the pr...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Editorial Department of Journal on Communications
2018-02-01
|
| Series: | Tongxin xuebao |
| Subjects: | |
| Online Access: | http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2018024/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841539454317101056 |
|---|---|
| author | Hengjun WANG Nianwen SI Yulong SONG Yidong SHAN |
| author_facet | Hengjun WANG Nianwen SI Yulong SONG Yidong SHAN |
| author_sort | Hengjun WANG |
| collection | DOAJ |
| description | LSTM and piecewise CNN were utilized to extract word vector features and global vector features,respectively.Then the two features were input to feed forward network for training.In model training,the probabilistic training method was adopted.Compared with the original dependency paring model,the proposed model focused more on global features,and used all potential dependency trees to update model parameters.Experiments on Chinese Penn Treebank 5 (CTB5) dataset show that,compared with the parsing model using LSTM or CNN only,the proposed model not only remains the relatively low model complexity,but also achieves higher accuracies. |
| format | Article |
| id | doaj-art-ba4a42abe53b4dd396e3a367b4102516 |
| institution | Kabale University |
| issn | 1000-436X |
| language | zho |
| publishDate | 2018-02-01 |
| publisher | Editorial Department of Journal on Communications |
| record_format | Article |
| series | Tongxin xuebao |
| spelling | doaj-art-ba4a42abe53b4dd396e3a367b41025162025-01-14T07:14:14ZzhoEditorial Department of Journal on CommunicationsTongxin xuebao1000-436X2018-02-0139536459716447Neural network model for dependency parsing incorporating global vector featureHengjun WANGNianwen SIYulong SONGYidong SHANLSTM and piecewise CNN were utilized to extract word vector features and global vector features,respectively.Then the two features were input to feed forward network for training.In model training,the probabilistic training method was adopted.Compared with the original dependency paring model,the proposed model focused more on global features,and used all potential dependency trees to update model parameters.Experiments on Chinese Penn Treebank 5 (CTB5) dataset show that,compared with the parsing model using LSTM or CNN only,the proposed model not only remains the relatively low model complexity,but also achieves higher accuracies.http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2018024/dependency parsinggraph-based modellong short-term memory networkconvolutional neural network,feature |
| spellingShingle | Hengjun WANG Nianwen SI Yulong SONG Yidong SHAN Neural network model for dependency parsing incorporating global vector feature Tongxin xuebao dependency parsing graph-based model long short-term memory network convolutional neural network,feature |
| title | Neural network model for dependency parsing incorporating global vector feature |
| title_full | Neural network model for dependency parsing incorporating global vector feature |
| title_fullStr | Neural network model for dependency parsing incorporating global vector feature |
| title_full_unstemmed | Neural network model for dependency parsing incorporating global vector feature |
| title_short | Neural network model for dependency parsing incorporating global vector feature |
| title_sort | neural network model for dependency parsing incorporating global vector feature |
| topic | dependency parsing graph-based model long short-term memory network convolutional neural network,feature |
| url | http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2018024/ |
| work_keys_str_mv | AT hengjunwang neuralnetworkmodelfordependencyparsingincorporatingglobalvectorfeature AT nianwensi neuralnetworkmodelfordependencyparsingincorporatingglobalvectorfeature AT yulongsong neuralnetworkmodelfordependencyparsingincorporatingglobalvectorfeature AT yidongshan neuralnetworkmodelfordependencyparsingincorporatingglobalvectorfeature |