基于Highway-BiLSTM网络的汉语谓语中心词识别研究

Ruizhang Huang; Wenfan Jin; Yanping Chen; Yongbin Qin; Qinghua Zheng

doi:10.11959/j.issn.1000-436x.2021027

基于Highway-BiLSTM网络的汉语谓语中心词识别研究

Translated title of the contribution: Research on Chinese predicate head recognition based on Highway-BiLSTM network

Ruizhang Huang
, Wenfan Jin
, Yanping Chen
, Yongbin Qin
, Qinghua Zheng

Faculty of Electronic and Information Engineering

Research output: Contribution to journal › Article › peer-review

3 Scopus citations

Abstract

Aiming at the problem of difficult recognition and uniqueness of Chinese predicate head, a Highway-BiLSTM model was proposed. Firstly, multi-layer BiLSTM networks were used to capture multi-granular semantic dependence in a sentence. Then, a Highway network was adopted to alleviate the problem of gradient disappearance. Finally, the output path was optimized by a constraint layer which was designed to guarantee the uniqueness of predicate head. The experimental results show that the proposed method effectively improves the performance of predicate head recognition.

Translated title of the contribution	Research on Chinese predicate head recognition based on Highway-BiLSTM network
Original language	Chinese (Traditional)
Pages (from-to)	100-107
Number of pages	8
Journal	Tongxin Xuebao/Journal on Communications
Volume	42
Issue number	1
DOIs	https://doi.org/10.11959/j.issn.1000-436x.2021027
State	Published - 25 Jan 2021

Access to Document

10.11959/j.issn.1000-436x.2021027

Cite this

@article{4f31c4d52822455ea986cc5d9de69bd4,

title = "基于Highway-BiLSTM网络的汉语谓语中心词识别研究",

abstract = "Aiming at the problem of difficult recognition and uniqueness of Chinese predicate head, a Highway-BiLSTM model was proposed. Firstly, multi-layer BiLSTM networks were used to capture multi-granular semantic dependence in a sentence. Then, a Highway network was adopted to alleviate the problem of gradient disappearance. Finally, the output path was optimized by a constraint layer which was designed to guarantee the uniqueness of predicate head. The experimental results show that the proposed method effectively improves the performance of predicate head recognition.",

keywords = "BiLSTM, Highway connection, Predicate head, Uniqueness",

author = "Ruizhang Huang and Wenfan Jin and Yanping Chen and Yongbin Qin and Qinghua Zheng",

year = "2021",

month = jan,

day = "25",

doi = "10.11959/j.issn.1000-436x.2021027",

language = "繁体中文",

volume = "42",

pages = "100--107",

journal = "Tongxin Xuebao/Journal on Communications",

issn = "1000-436X",

publisher = "Editorial Board of Journal on Communications",

number = "1",

}

TY - JOUR

T1 - 基于Highway-BiLSTM网络的汉语谓语中心词识别研究

AU - Huang, Ruizhang

AU - Jin, Wenfan

AU - Chen, Yanping

AU - Qin, Yongbin

AU - Zheng, Qinghua

PY - 2021/1/25

Y1 - 2021/1/25

N2 - Aiming at the problem of difficult recognition and uniqueness of Chinese predicate head, a Highway-BiLSTM model was proposed. Firstly, multi-layer BiLSTM networks were used to capture multi-granular semantic dependence in a sentence. Then, a Highway network was adopted to alleviate the problem of gradient disappearance. Finally, the output path was optimized by a constraint layer which was designed to guarantee the uniqueness of predicate head. The experimental results show that the proposed method effectively improves the performance of predicate head recognition.

AB - Aiming at the problem of difficult recognition and uniqueness of Chinese predicate head, a Highway-BiLSTM model was proposed. Firstly, multi-layer BiLSTM networks were used to capture multi-granular semantic dependence in a sentence. Then, a Highway network was adopted to alleviate the problem of gradient disappearance. Finally, the output path was optimized by a constraint layer which was designed to guarantee the uniqueness of predicate head. The experimental results show that the proposed method effectively improves the performance of predicate head recognition.

KW - BiLSTM

KW - Highway connection

KW - Predicate head

KW - Uniqueness

UR - https://www.scopus.com/pages/publications/85100323420

U2 - 10.11959/j.issn.1000-436x.2021027

DO - 10.11959/j.issn.1000-436x.2021027

M3 - 文章

AN - SCOPUS:85100323420

SN - 1000-436X

VL - 42

SP - 100

EP - 107

JO - Tongxin Xuebao/Journal on Communications

JF - Tongxin Xuebao/Journal on Communications

IS - 1

ER -

基于Highway-BiLSTM网络的汉语谓语中心词识别研究

Abstract

Access to Document

Other files and links

Fingerprint

Cite this