Mention Detection for Improving Coreference Resolution in Russian Texts: A Machine Learning Approach

被引：0
作者：
Toldova, Svetlana [1 ]
Ionov, Max [2 ]
机构：
[1] Higher Sch Econ, Natl Res Inst, Moscow, Russia
[2] Moscow MV Lomonosov State Univ, Moscow, Russia
来源：
COMPUTACION Y SISTEMAS | 2016年 / 20卷 / 04期
基金：
俄罗斯基础研究基金会;
关键词：
Coreference resolution; discourse-new detection; singleton detection; discourse processing; natural language processing; machine learning;
D O I：
10.13053/CyS-20-4-2480
中图分类号：
TP [自动化技术、计算机技术];
学科分类号：
0812 ;
摘要：
Coreference resolution task is a well-known NLP application that was proven helpful for all high-level NLP applications: machine translation, summarization, and others. Mention detection is the sub-task of detecting the discourse status of each noun phrase, classifying it as a discourse-new, singleton (mentioned only once) or discourse-old occurrence. It has been shown that this task applied to a coreference resolution system may increase its overall performance. So, we decided to adapt current approaches for English language into Russian. We present some quality results of experiments regarding classifiers for mention detection and their application into the coreference resolution task in Russian languages.
引用
收藏
页码：681 / 696
页数：16
相关论文
共 38 条
  • [1] Arutyunova N., 1980, NOMINATION GEN QUEST
  • [2] Batista G.E.A.P.A. Bazzan A.L.C. Monard M.C, 2003, BALANCING TRAINING D, P10
  • [3] Bean D.L., 1999, P 37 ANN M ASS COMP, P373, DOI [10.3115/1034678.1034737, DOI 10.3115/1034678.1034737]
  • [4] Bonch-Osmolovskaya A., 2012, INTRO NOUN PHRASES C
  • [5] Gelbukh A., 2002, COMPUT SIST, V5, P204
  • [6] Givon Talmy., 1983, TOPIC CONTINUITY DIS
  • [7] Grosz B. J., 1986, Computational Linguistics, V12, P175
  • [8] GROSZ BJ, 1995, COMPUT LINGUIST, V21, P203
  • [9] Hawkins J. A., 1978, DEFINITENESS INDEFMI
  • [10] Hirschman L., 1998, 7 MESSAGE UNDERSTAND
1234