信息抽取研究综述

信息抽取研究旨在为人们提供更有力的信息获取工具,以应对信息爆炸带来的严重挑战。与信息检索不同,信息抽取直接从自然语言文本中抽取事实信息。过去十多年来,信息抽取逐步发展成为自然语言处理领域的一个重要分支,其独特的发展轨迹———通过系统化、大规模的定量评测推动研究向前发展,以及某些成功启示,如部分分析技术的有效性、快速NLP系统开发的必要性,都极大地推动了自然语言处理研究的发展,促进了NLP研究与应用的紧密结合。回顾信息抽取研究的历史,总结信息抽取研究的现状,将有助于这方面研究工作向前发展。

Research on Information Extraction:A Survey

The research on Information Extraction aims at providing more powerful information access tools to help peo-ple overcome the problem of information overloading.Unlike Information Retrieval,Information Extraction Systems extract factual information directly from natural language texts.In the last decade,Information Extraction has become an impor-tant sub-field of Natural Language Processing.Its unique development track,i.e.accelerating research via systematical and large scale evaluation,and some successful experience,such as the effectiveness of partial-parsing techniques and the importance of fast development cycles,have made it a great and most important impetus to the research of NLP in the last decade.Moreover,Information Extraction has built a more effective connection between NLP researchers and NLP system developers.It will be helpful to review the history and investigate the state of the art of Information Extraction.