Leveraging Linguistic Structures for Named Entity Recognition with Bidirectional Recursive Neural Networks
碩士 === 國立臺灣大學 === 資訊工程學研究所 === 106 === Named Entity Recognition (NER) is an important task which locates proper names in text for downstream tasks, e.g. to facilitate natural language understanding. The problem is often casted from structured prediction of text chunks to sequential labeling of token...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | en_US |
Published: |
2017
|
Online Access: | http://ndltd.ncl.edu.tw/handle/xbpj78 |
id |
ndltd-TW-106NTU05392017 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-106NTU053920172019-05-16T00:22:53Z http://ndltd.ncl.edu.tw/handle/xbpj78 Leveraging Linguistic Structures for Named Entity Recognition with Bidirectional Recursive Neural Networks 利用語法結構之雙向遞迴類神經網路於命名實體辨識之研究 Peng-Hsuan Li 李朋軒 碩士 國立臺灣大學 資訊工程學研究所 106 Named Entity Recognition (NER) is an important task which locates proper names in text for downstream tasks, e.g. to facilitate natural language understanding. The problem is often casted from structured prediction of text chunks to sequential labeling of tokens. Such sequential approaches have achieved high performance with models like conditional random fields and recurrent neural networks. However, named entities should be linguistic constituents, and sequential token labeling neglects this information. In the thesis, we propose a constituency-oriented approach which fully utilizes linguistic structures in text. First, to leverage the prior knowledge of hierarchical phrase structures, we generate parses and alter them into constituency graphs that minimize inconsistencies between parses and named entities. Then, we use Bidirectional Recursive Neural Networks (BRNN) to propagate relevant structure information to each constituent. We use a bottom-up pass to capture the local information and a top-down pass to capture the global information. Experiments show that this approach is comparable to sequential token labeling, and significant improvements can be seen on OntoNotes 5.0 NER, with F1 scores over 87\%. Jane Yung-jen Hsu 許永真 2017 學位論文 ; thesis 74 en_US |
collection |
NDLTD |
language |
en_US |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 國立臺灣大學 === 資訊工程學研究所 === 106 === Named Entity Recognition (NER) is an important task which locates proper names in text for downstream tasks, e.g. to facilitate natural language understanding. The problem is often casted from structured prediction of text chunks to sequential labeling of tokens. Such sequential approaches have achieved high performance with models like conditional random fields and recurrent neural networks. However, named entities should be linguistic constituents, and sequential token labeling neglects this information.
In the thesis, we propose a constituency-oriented approach which fully utilizes linguistic structures in text. First, to leverage the prior knowledge of hierarchical phrase structures, we generate parses and alter them into constituency graphs that minimize inconsistencies between parses and named entities. Then, we use Bidirectional Recursive Neural Networks (BRNN) to propagate relevant structure information to each constituent. We use a bottom-up pass to capture the local information and a top-down pass to capture the global information. Experiments show that this approach is comparable to sequential token labeling, and significant improvements can be seen on OntoNotes 5.0 NER, with F1 scores over 87\%.
|
author2 |
Jane Yung-jen Hsu |
author_facet |
Jane Yung-jen Hsu Peng-Hsuan Li 李朋軒 |
author |
Peng-Hsuan Li 李朋軒 |
spellingShingle |
Peng-Hsuan Li 李朋軒 Leveraging Linguistic Structures for Named Entity Recognition with Bidirectional Recursive Neural Networks |
author_sort |
Peng-Hsuan Li |
title |
Leveraging Linguistic Structures for Named Entity Recognition with Bidirectional Recursive Neural Networks |
title_short |
Leveraging Linguistic Structures for Named Entity Recognition with Bidirectional Recursive Neural Networks |
title_full |
Leveraging Linguistic Structures for Named Entity Recognition with Bidirectional Recursive Neural Networks |
title_fullStr |
Leveraging Linguistic Structures for Named Entity Recognition with Bidirectional Recursive Neural Networks |
title_full_unstemmed |
Leveraging Linguistic Structures for Named Entity Recognition with Bidirectional Recursive Neural Networks |
title_sort |
leveraging linguistic structures for named entity recognition with bidirectional recursive neural networks |
publishDate |
2017 |
url |
http://ndltd.ncl.edu.tw/handle/xbpj78 |
work_keys_str_mv |
AT penghsuanli leveraginglinguisticstructuresfornamedentityrecognitionwithbidirectionalrecursiveneuralnetworks AT lǐpéngxuān leveraginglinguisticstructuresfornamedentityrecognitionwithbidirectionalrecursiveneuralnetworks AT penghsuanli lìyòngyǔfǎjiégòuzhīshuāngxiàngdìhuílèishénjīngwǎnglùyúmìngmíngshítǐbiànshízhīyánjiū AT lǐpéngxuān lìyòngyǔfǎjiégòuzhīshuāngxiàngdìhuílèishénjīngwǎnglùyúmìngmíngshítǐbiànshízhīyánjiū |
_version_ |
1719165176408702976 |