Development of Intelligent Full-Text Information Processing System Based on Efficient Pattern Matching Algorithms

本课题的目标是开发一种利用模式匹配和机器学习技术的新型文本数据库管理系统。今年我们成功完成了项目，取得了以下成果：(1)实现并评估了智能全文处理系统。(2)针对几种压缩方法，开发了高效的压缩文本模式匹配算法。(3) 1997年6月，我们在丹麦召开的组合模式匹配（CPM97）会议上提出了上述算法，并与该领域的专家进行了讨论。(4)开发了一种高效的图像和通配符匹配算法。(5)我们开发了文本处理学习系统，并将其整合到(1)的系统中。

The goal of this project is to develop a new type text database management system which exploits techniques of pattern matching and machine learning. We completed the project successfully with the following results achieved in this year.(1) We implemented and evaluated an intelligent full-text processing system.(2) We developed efficient pattern matching algorithms in compressed texts for several compression methods.(3) We presented the above algorithms at conference of Combinatorial Pattern Matching (CPM97) in Denmark at June 1997, and discussed them with experts in this research area.(4) We developed an efficient pattern matching algorithm which deals with pictures and wildcards.(5) We developed learning system for text processing and then incorporated it into the system of (1).