With the development of network technology, the number of online texts is increasing at an exponential speed. How to acquire useful information accurately and efficiently becomes the focus of researchers, which accelerates the development of text processing. Automatic collection and liquidation of required information become a challenge and a chance of development. And information retrieval technology is the key to the solution of these problems. According to different backgrounds and application purpose, information retrieval technology has evolved into several subareas: text retrieval, text filtering, text categorization, text abstraction, and question answering systems, etc.Since the major form of information is text, we discuss Chinese text information retrieval and relevant issues in this thesis. The main object of text retrieval study is large-scale, non-structured real-world texts for categorization, retrieval, filtering, and abstraction to meet user's requirement.Issues discussed in this thesis include indexing technology for improving retrieval efficiency and relevance feedback for improving system precision.Text categorization can be used as independent application system and as a component for an information retrieval system for or improving retrieval efficiency. The propose that a hierarchical method for Chinese text categorization be used in a retrieval system. |