UM  > 科技學院  > 電腦及資訊科學系
Graph-based lexicon regularization for PCFG with latent annotations
Zeng X.2; Wong D.F.2; Chao L.S.2; Trancoso I.1
2015-03-01
Source PublicationIEEE Transactions on Audio, Speech and Language Processing
ISSN15587916
Volume23Issue:3Pages:441-450
Abstract

This paper aims at learning a better probabilistic context-free grammar with latent annotations (PCFG-LA) by using a graph propagation (GP) technique. We propose leveraging the GP to regularize the lexical model of the grammar. The proposed approach constructs $k$-nearest neighbor ($k$ -NN) similarity graphs over words with identical pre-terminal (part-of-speech) tags, for propagating the probabilities of latent annotations given the words. The graphs demonstrate the relationship between words in syntactic and semantic levels, estimated by using a neural word representation method based on Recursive autoencoder (RAE). We modify the conventional PCFG-LA parameter estimation algorithm, expectation maximization (EM), by incorporating a GP process subsequent to the M-step. The GP encourages the smoothness among the graph vertices, where different words under similar syntactic and semantic environments should have approximate posterior distributions of nonterminal subcategories. The proposed PCFG-LA learning approach was evaluated together with a hierarchical split-and-merge training strategy, on parsing tasks for English, Chinese and Portuguese. The empirical results reveal two crucial findings: 1) regularizing the lexicons with GP results in positive effects to parsing accuracy; and 2) learning with unlabeled data can also expand the PCFG-LA lexicons.

KeywordGraph Propagation Natural Language Processing Neural Word Representation Syntax Parsing
DOI10.1109/TASLP.2015.2389034
URLView the original
Indexed BySCI
Language英语
WOS Research AreaAcoustics ; Engineering
WOS SubjectAcoustics ; Engineering, Electrical & Electronic
WOS IDWOS:000350876100003
全文获取链接
引用统计
被引频次[WOS]:2   [WOS记录]     [WOS相关记录]
Document TypeJournal article
专题DEPARTMENT OF COMPUTER AND INFORMATION SCIENCE
Affiliation1.Instituto de Engenharia de Sistemas e Computadores Investigação e Desenvolvimento em Lisboa
2.Universidade de Macau
推荐引用方式
GB/T 7714
Zeng X.,Wong D.F.,Chao L.S.,et al. Graph-based lexicon regularization for PCFG with latent annotations[J]. IEEE Transactions on Audio, Speech and Language Processing,2015,23(3):441-450.
APA Zeng X.,Wong D.F.,Chao L.S.,&Trancoso I..(2015).Graph-based lexicon regularization for PCFG with latent annotations.IEEE Transactions on Audio, Speech and Language Processing,23(3),441-450.
MLA Zeng X.,et al."Graph-based lexicon regularization for PCFG with latent annotations".IEEE Transactions on Audio, Speech and Language Processing 23.3(2015):441-450.
个性服务
推荐该条目
保存到收藏夹
查看访问统计
导出为Endnote文件
Google Scholar
中相似的文章 Google Scholar
[Zeng X.]的文章
[Wong D.F.]的文章
[Chao L.S.]的文章
Baidu academic
中相似的文章 Baidu academic
[Zeng X.]的文章
[Wong D.F.]的文章
[Chao L.S.]的文章
Bing Scholar
中相似的文章 Bing Scholar
[Zeng X.]的文章
[Wong D.F.]的文章
[Chao L.S.]的文章
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。