Abstract: The paper studies the related technologies of the Chinese text knowledge discovery on Internet, according to the features of search engines classification, comparing and analyzing search engines with features of text classification, discovers that the engines of simplex search text information is more suitable for text classification. This paper proposes a Internet achieving information prototype based on text classification, proposes an extraction method on user knowledge of interest, users can extract text information of interest to improve accuracy of text information mining.


Key words: knowledge discovery, text classification, search engine, rule mining