Based on Conditional Random Fields（CRF）, the article identifies the Chinese prepositional phrase. In order to identify the prepositional phrase effectively, the article counts and analyzes the external and internal linguistic features of the prepositional phrase. The prepositional phrases without nesting and nested prepositional phrases are identified with the complex feature model,and in open tests,the best F value can reach 90.29% and 89.99% respectively.
朱丹浩 王东波 谢靖. 基于条件随机场的介宾结构自动识别*[J]. 现代图书情报技术, 2010, 26(7/8): 79-83.
Zhu Danhao Wang Dongbo Xie Jing. Automatic Identification of Prepositional Phrase Based on Conditional Random Field. New Technology of Library and Information Service, 2010, 26(7/8): 79-83.
［1］ 周强,孙茂松,黄昌宁.汉语句子的组块分析体系［J］.计算机学报,1999,22(11):1158-1165.
［2］ Brill E,Resnik P.A Rule-based Approach to Prepositional Phrase-attachment Disambiguation［C］. In:Proceedings of the 15th Conference on Computational Linguistics,Kyoto,Japan.1994.
［3］ Pantel P,Lin D.An Unsupervised Approach to Prepositional Phrase Attachment Using Contextually Similar Words［C］.In:Proceedings of Association for Computational Linguistics.1998.
［4］ Schwartz L,Aikawa T,Quirk C.Disambiguation of English PP Attachment Using Multilingual Aligned Data［C］.In: Proceedings of MT Summit IX,New Orleans,USA.2003.
［5］ McLauchlan M.Thesauruses for Prepositional Phrase Attachment［C］.In:Proceedings of CoNLL,Boston,MA,USA.2004:73-80.
［6］ 干俊伟,黄德根.汉语介词短语的自动识别［J］.中文信息学报,2005,19(4):17-23.
［7］ 奚建清，罗强.基于HMM的汉语介词短语自动识别研究［J］.计算机工程,2007,33(3):172-173,182.
［8］ 于俊涛. 基于最大熵的汉语介词短语自动识别［D］.大连:大连理工大学,2006:35.
［9］ The Features of CRF++［EB/OL］.［2009-12-06］.http://crfpp.sourceforge.net/#features.
［10］ Lafferty J, McCallum A, Pereira F. Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data［C］. In: Proceedings of International Conference on Machine Learning.2001:3-4.
［11］ Sutton C, McCallum A. An Introduction to Conditional Random Fields for Relational Learning［M］. MIT Press, 2006:2-7.
