中文信息提取技术研究

As information extraction (IE) recently becoming more and more regarded in the world wide, the necessity and conditions of studying Chinese information have been possessed. This research project is intended to establish a fundamental set of Chinese IE technologies based on the study of existing relevant research results and the rich grammatical and semantic knowledge of the Chinese language already presented, as well as to implement an applicable Chinese IE system prototype for the test and improvement of the technologies. Most of the outcome has been applied to a large-scale system of Web page collection and information extraction (namely, the PKU Fame system), with the expected useful effects. More further study on the various aspects of Chinese IE technology would also be helpful and necessary for its wider application and more profound development.

信息提取技术是近年来在国际上备受关注的综合性语言信息处理技术，本项研究对汉语自然语言处理技术的综合应用及汉语语法、语义分析理论都有着积极重要的意义。研究内容包括针对汉语信息提取的综合性语言知识库、汉语文本浅层分析技术、通用的信息描述模式和通用的中文信息提取系统。本课题文理结合特点显著，理论开拓与工程实践紧密结合。