基于半监督CRF的缩略词扩展解释识别

doi:10.3969/j.issn.1000-3428.2013.04.047

计算机工程 ›› 2013, Vol. 39 ›› Issue (4): 203-209. doi: 10.3969/j.issn.1000-3428.2013.04.047

基于半监督CRF的缩略词扩展解释识别

陈季梦^a，刘杰^a，黄亚楼^a,b，刘天笔^b，刘才华^a

(南开大学 a. 信息技术科学学院；b. 软件学院，天津 300071)

收稿日期:2012-05-17 出版日期:2013-04-15 发布日期:2013-04-12
作者简介:陈季梦(1987－)，女，博士研究生，主研方向：机器学习，自然语言处理；刘杰，副教授、博士；黄亚楼，教授、博士生导师；刘天笔，硕士研究生；刘才华，博士研究生
基金资助:
国家自然科学基金资助项目(61105049)；高等学校博士学科点专项科研基金资助项目(20100031110096)；中央高校基本科研业务费专项基金资助项目(65010571)

Abbreviation Expansion Interpretation Recognition Based on Semi-supervised CRF

CHEN Ji-meng ^a, LIU Jie ^a, HUANG Ya-lou ^a,b, LIU Tian-bi ^b, LIU Cai-hua ^a

(a. College of Information Technical Science; b. Software College, Nankai University, Tianjin 300071, China)

Received:2012-05-17 Online:2013-04-15 Published:2013-04-12

摘要/Abstract

摘要： 缩略词拓展解释识别任务中标注样本较少，无法从中总结出全面的规则或采用有监督的学习方法来学习。为此，提出一种基于半监督条件随机场(CRF)的缩略词扩展解释识别模型，利用广泛的未标注样本和较少的标注样本寻找序列文本中恰当的语句，以解释给定的缩略词。使用较少的标注序列样本训练一个全监督CRF模型，针对未标注序列样本，采用最小序列熵学习样本之间的联系，结合标注样本和未标注样本，利用半监督自学习方法学习两者的关系。实验结果表明，该模型的序列F1值达到84.73%，高于支持向量机和全监督CRF基准算法。

关键词: 扩展解释, 半监督, 条件随机场, 序列熵, 序列标注

Abstract: Because lack of labeled samples in acronym expansion recognition task, rule-based or supervised methods are unsuitable to find a proper phrase in sequential texts to explain a given acronym. A semi-supervised Conditional Random Field(CRF) is designed to make use of sufficient unlabeled samples and a few labeled samples. A supervised CRF model by a few labeled sequential samples is trained. Minimum sequential entropy is employed to learn relationship between samples without labels. A self training model is applied to analyzes the relationship between labeled and unlabeled samples. Experimental results show this model achieves the best performance of sequential F1(84.73%) against the state-of-the-art baselines including Support Vector Machine(SVM) and supervised CRF.

Key words: expansion interpretation, semi-supervised, conditional random field, sequence entropy, sequence labeling

中图分类号:

TP391

陈季梦, 刘杰, 黄亚楼, 刘天笔, 刘才华. 基于半监督CRF的缩略词扩展解释识别[J]. 计算机工程, 2013, 39(4): 203-209.

CHEN Ji-Meng, LIU Jie, HUANG E-Lou, LIU Tian-Bi, LIU Cai-Hua. Abbreviation Expansion Interpretation Recognition Based on Semi-supervised CRF[J]. Computer Engineering, 2013, 39(4): 203-209.

http://www.ecice06.com/CN/Y2013/V39/I4/203

参考文献

[1] Lafferty J, Mccallum A, Fernando P. Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data[C]//Proc. of International Conference on Machine Learning. [S. l.]: AAAI Press, 2001.
[2] Kazem T, Gilbreth J. Recognizing Acronyms and Their Definitions[J]. International Journal on Document Analysis and Recognition, 1999, 1(4): 191-198.
[3] Stuart Y. Automatic Extraction of Acronyms from Text[C]//Proc. of New Zealand Computer Science Research Students’ Conference. Waikato, New Zealand: [s. n.], 1999.
[4] Sanchez D, Isern D. Automatic Extraction of Acronym Definitions from the Web[J]. Applied Intelligence Journal, 2011, 34(2): 311-327.
[5] 朱江涛, 蔡东风, 张桂平. 一种基于网络的英文缩略语信息的自动抽取方法[C]//全国第8届计算语言学联合学术会议论文集. 南京: [出版者不详], 2005.
[6] Chang J T, Schütze H, Altman R B. Creating an Online Dictionary of Abbreviations from MEDLINE[J]. Journal of the American Medical Informatics Association, 2002, 9(6): 612-620.
[7] Nadeau D, Peter D T. A Supervised Learning Approach to Acronym Identification[C]//Proc. of the 8th Canadian Conference on Artificial Intelligence. Ottawa, Canada: IEEE Press, 2005.
[8] Xu Jun, Huang Yalou. Using SVM to Extract Acronyms from Text[J]. Soft Computing, 2006, 11(4): 369-373.
[9] 高永梅, 黄亚楼, 倪维健, 等. 基于排序支持向量机的缩略词自动提取方法[J]. 模式识别与人工智能, 2008, 21(2): 186-192.
[10] Liu Jie, Chen Jimeng, Zhang Yi, et al. Learning Conditional Random Fields with Latent Sparse Features for Acronym Expansion Finding[J]//Proc. of the 20th ACM Conference on Information and Knowledge Management. Glasgow, UK: ACM Press, 2011.
[11] Rabiner L R. A Tutorial on Hidden Markov Models and Selected Applications in Speech Recognition[J]. Proceedings of the IEEE, 1989, 77(2): 257-286.
[12] Tasker B, Abbeel P, Daphne K. Discriminative Probabilistic Models for Relational Data[C]//Proc. of the 18th Annual Conference on Uncertainty in Arti?cial Intelligence. San Francisco, USA: Morgan Kaufmann Publishers, 2002.
[13] Zhu Xiaojin, Andrew B G. Introduction to Semi-supervised Learning[J]. Synthesis Lectures on Artificial Intelligence and Machine Learning, 2009, 3(1): 1-130.
[14] Mann G S, McCallum A. Efficient Computation of Entropy Gradient for Semi-supervised Conditional Random Fields[C]//Proc. of the 2007 Conference of the North American Chapter of the Association for Computational Linguistics. Stroudsburg, USA: Association for Computational Linguistics, 2007.
[15] Jiao Feng, Wang Shaojun, Lee Chi-Hoon, et al. Semi- supervised Conditional Random Fields for Improved Sequence Segmentation and Labeling[C]//Proc. of the 21st International Conference on Computational Linguistics and the 44th Annual Meeting of the Association for Computational Linguistics. Sydney, Australia: Association for Computational Linguistics, 2006.
编辑刘冰
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

选择文件类型/文献管理软件名称

选择包含的内容

基于半监督CRF的缩略词扩展解释识别

Abbreviation Expansion Interpretation Recognition Based on Semi-supervised CRF

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	陈仲磊, 伊鹏, 陈祥, 胡涛. 基于集成学习的系统调用实时异常检测框架[J]. 计算机工程, 2023, 49(6): 162-169,179.
[2]	司逸晨, 管有庆. 基于Transformer编码器的中文命名实体识别模型[J]. 计算机工程, 2022, 48(7): 66-72.
[3]	王刚, 孙媛媛, 陈彦光, 林鸿飞. 面向法律文书的分段式摘要模型[J]. 计算机工程, 2022, 48(6): 288-294.
[4]	李军怀, 陈苗苗, 王怀军, 崔颖安, 张爱华. 基于ALBERT-BGRU-CRF的中文命名实体识别方法[J]. 计算机工程, 2022, 48(6): 89-94,106.
[5]	王树芬, 张哲, 马士尧, 陈俞强, 伍一. 一种鲁棒的半监督联邦学习系统[J]. 计算机工程, 2022, 48(6): 107-114,123.
[6]	佘朝阳, 严馨, 徐广义, 陈玮, 邓忠莹. 融合数据增强与半监督学习的药物不良反应检测[J]. 计算机工程, 2022, 48(6): 314-320.
[7]	崔丽平, 古丽拉·阿东别克, 王智悦. 基于有向图模型的旅游领域命名实体识别[J]. 计算机工程, 2022, 48(2): 306-313.
[8]	胡彬, 王晓军, 张雷. 一种半监督对抗鲁棒模型无关元学习方法[J]. 计算机工程, 2022, 48(12): 112-118.
[9]	高伟, 吴顺. 基于多尺度注意力半监督学习的老照片划痕修复[J]. 计算机工程, 2022, 48(10): 245-251,261.
[10]	冯桫, 刘井平, 蒋海云, 肖仰华. 基于机器阅读理解模型与众包验证的属性值抽取方法[J]. 计算机工程, 2021, 47(5): 97-103.
[11]	康璐璐, 范兴容, 王茜竹, 杨晓雅, 明蕊. 基于特征组分层与半监督学习的鼠标轨迹识别[J]. 计算机工程, 2021, 47(4): 277-284.
[12]	薛子晗, 潘迪, 何丽. 结合改进密度峰值聚类的LGC半监督学习方法优化[J]. 计算机工程, 2021, 47(2): 77-83,89.
[13]	吕江海, 杜军平, 周南, 薛哲. 基于膨胀卷积迭代与注意力机制的实体名识别方法[J]. 计算机工程, 2021, 47(1): 58-65,71.
[14]	何阳宇, 晏雷, 易绵竹, 李宏欣. 融合CRF与规则的老挝语军事领域命名实体识别方法[J]. 计算机工程, 2020, 46(8): 297-304.
[15]	杨飘, 董文永. 基于BERT嵌入的中文命名实体识别方法[J]. 计算机工程, 2020, 46(4): 40-45,52.

模态框（Modal）标题

选择文件类型/文献管理软件名称

选择包含的内容

基于半监督CRF的缩略词扩展解释识别

Abbreviation Expansion Interpretation Recognition Based on Semi-supervised CRF

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价