A Survey of Paraphrasing and Textual Entailment Methods

被引:169
作者
Androutsopoulos, Ion [1 ]
Malakasiotis, Prodromos [1 ]
机构
[1] Athens Univ Econ & Business, Dept Informat, GR-10434 Athens, Greece
关键词
INFORMATION EXTRACTION; SENTENCE COMPRESSION; CONSTRUCTION; CORPUS;
D O I
10.1613/jair.2985
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Paraphrasing methods recognize, generate, or extract phrases, sentences, or longer natural language expressions that convey almost the same information. Textual entailment methods, on the other hand, recognize, generate, or extract pairs of natural language expressions, such that a human who reads (and trusts) the first element of a pair would most likely infer that the other element is also true. Paraphrasing can be seen as bidirectional textual entailment and methods from the two areas are often similar. Both kinds of methods are useful, at least in principle, in a wide range of natural language processing applications, including question answering, summarization, text generation, and machine translation. We summarize key ideas from the two areas by considering in turn recognition, generation, and extraction methods, also pointing to prominent articles and resources.
引用
收藏
页码:135 / 187
页数:53
相关论文
共 229 条
  • [1] ANDROUTSOPOULOS I, 2007, NAT LANG ENG, V13, P291
  • [2] Androutsopoulos I., 1995, Natural Lang. Eng., V1, P29, DOI DOI 10.1017/S135132490000005X
  • [3] [Anonymous], 2006, Proceedings of the human language technology conference of the NAACL, companion volume: Short papers
  • [4] [Anonymous], 2006, P 21 INT C COMP LING
  • [5] [Anonymous], P 2 INT WORKSH PAR A
  • [6] [Anonymous], OXFORD HDB COMPUTATI
  • [7] [Anonymous], 2003, OXFORD HDB COMP LING
  • [8] [Anonymous], 2006, P HUMAN LANGUAGE TEC
  • [9] [Anonymous], 2004, Introduction to Machine Learning
  • [10] [Anonymous], P 3 INT WORKSH PAR I