Retrieval methods of natural language based on automatic indexing

Abstract

Since natural language enter the computer retrieval system, due to the natural language retrieval is not restricted by professional experience, knowledge background, retrieval experience by users, and above reasons favored by the users. As the title of the Chinese literature is the concentrated reflection of Chinese literature content, it reflects the central idea of the literature. Retrieval methods of natural language described in this article is limited to literature title in subject indexing. The basic idea of this method is, with automatic indexing methods respectively the literature title in the database of retrieval system used in natural language retrieval for automatic word indexing. To control the concept of a given keyword, namely meaning transformation, form the final indexing words. Then, using the vector space model for the index data in the database will be “or” operation to retrieve, forming a document set B. For each document title in set B for automatic indexing, the title of each article for automatic indexing, indexing terms for the formation and retrieval of natural language indexing terms similarity calculation, sorted according to similarity of each document in set B. The first best match the requirements presented to the user documentation. This method is a simple and practical method of natural language retrieval.