Cross-Domain and Semisupervised Named Entity Recognition in Chinese Social Media: A Unified Model

Cross-Domain and Semisupervised Named Entity Recognition in Chinese Social Media: A Unified Model
复制标题

中国社交媒体中的跨域半监督命名实体识别:统一模型

DOI:
10.1109/taslp.2018.2856625
复制
发表时间:
2018-11
影响因子:
5.4
通讯作者:
Li Sujian
Li Sujian
中科院分区:
计算机科学2区
文献类型:
--
作者:
Xu Jingjing;He Hangfeng;Sun Xu;Ren Xuancheng;Li Sujian

文献摘要

参考文献

被引文献

相似文献

中文社交媒体中的命名实体识别(NER)是一项重要但具有挑战性的任务,因为中文社交媒体语言是非正式的和嘈杂的。以往的NER方法大多集中在领域内的监督学习上,但受限于社交媒体中稀缺的标注数据。在本文中,我们提出了结合形式领域中的大量语料库和大量的未注释文本来提高社交媒体中的NER性能。我们提出了一个可以从域外语料库和域内未标注文本中学习的统一模型。统一模型由两部分组成。一种是跨域学习,另一种是半监督学习。跨域学习可以基于领域相似度学习域外信息。半监督学习通过自我训练来学习领域内未标注的信息。实验结果表明,我们的统一模型比强基线提高了9.57%,达到了最先进的性能。
Named entity recognition (NER) in Chinese social media is an important, but challenging task because Chinese social media language is informal and noisy. Most previous methods on NER focus on in-domain supervised learning, which is limited by scarce annotated data in social media. In this paper, we present that sufficient corpora in formal domains and massive unannotated text can be combined to improve the NER performance in social media. We propose a unified model which can learn from out-of-domain corpora and in-domain unannotated text. The unified model is composed of two parts. One is for cross-domain learning and the other is for semisupervised learning. Cross-domain learning can learn out-of-domain information based on domain similarity. Semisupervised learning can learn in-domain unannotated information by self-training. Experimental results show that our unified model yields a 9.57% improvement over strong baselines and achieves the state-of-the-art performance.1
DOI: --
发表时间: 2009-07
期刊: --
影响因子: --
作者:
Xuan Sun;Takuya Matsuzaki;Daisuke Okanohara;Junichi Tsujii
通讯作者: Xuan Sun;Takuya Matsuzaki;Daisuke Okanohara;Junichi Tsujii
DOI: 10.18653/v1/e17-2113
发表时间: 2016-11
期刊: ArXiv
影响因子: --
作者:
Hangfeng He;Xu Sun
通讯作者: Hangfeng He;Xu Sun
DOI: --
发表时间: 2007-06
期刊: --
影响因子: --
作者:
Rebecca Watson;Ted Briscoe
通讯作者: Rebecca Watson;Ted Briscoe
DOI: --
发表时间: 2009-12
期刊: --
影响因子: --
作者:
N. Ye;Wee Sun Lee;Hai Leong Chieu;Dan Wu
通讯作者: N. Ye;Wee Sun Lee;Hai Leong Chieu;Dan Wu
DOI: 10.3115/v1/n15-1075
发表时间: 2015
期刊: --
影响因子: --
作者:
Colin Cherry;Hongyu Guo
通讯作者: Colin Cherry;Hongyu Guo