自动文摘系统评测方法的回顾与展望  

The Overview and Prospect of Automatic Summarization Evaluation

作　　者：刘茵[1] 李弼程[1] 

机构地区：[1]信息工程大学信息工程学院信息科学系,郑州450002

出　　处：《情报学报》2008年第2期235-243,共9页Journal of the China Society for Scientific and Technical Information

摘　　要：自动文摘系统的评测是自动文摘技术研究与发展的关键部分。随着摘要技术的快速发展，国外学者十分重视评测技术，研究工作也如火如荼，而国内相关工作相对贫乏。文摘评测对文摘系统的性能有重大影响，对自然语言处理、信息检索等相关技术有着促进作用。本文详细介绍了文摘评测的两大方法：内部评测和外部评测，并对多年来文摘评测中应用的关键技术做了简要总结，最后指出文摘评测面临的挑战及未来需要努力的方向。The summarization evaluation is a key part of the research and development of the summarization technology. As the rapid development of the summarization technology, researchers pay more attention to the evaluation of summarization. Evaluation has a great effect on the summarization system, which can improve the performance of system and offer many advantages to the relative subjects such as NLP and IR. Both intrinsic and extrinsic approaches are discussed. The key techniques and criteria applied in evaluation are briefly summarized. At last, the challenge and future direction of evaluation are indicated.

关 键 词：自动文摘 评测方法 系统摘要 内部评测 外部评测 

分 类 号：G356.7[文化科学—情报学] TP391.41[自动化与计算机技术—计算机应用技术]