Method for classifying Chinese webpages based on keyword frequency analysis

Abstract

The invention relates to a method for classifying Chinese webpages based on keyword frequency analysis. According to the analyzed keywords of the Chinese webpages and a Chinese classification subject thesaurus, the fuzzy matching of the classification of the Chinese webpages is carried out; and through the obtained HTML source code of the webpage, the webpage is pretreated. Through the testing and analysis, a regular expression filter is utilized to filter noise information; a Chinese text of the webpage is extracted; then, through a word classifier and a keyword frequency analyzer, the extracted Chinese text information is subjected to word classification; through the weighed ranking of the word in the text and fuzzy classification algorithm of the webpage, the class ranking of the class which the webpage keyword belongs to is obtained; and the keywords ranking in the several tops are selected and subjected to calculation of membership rate to obtain the fuzzy matching result of the class which the webpage belongs to. The method is favorable for organizing mass information on the Internet with high efficiency and is used for interestingness analysis of Internet users, catalogue updating of search engines, mining of Web contents, online document management, and digital library construction.

Description

technical field [0001] The present invention is aimed at the research of the keyword frequency analysis of Chinese webpage and the webpage classification method based on the keyword frequency analysis, and mainly studies how to filter and extract the content of the Chinese webpage through technical means, word segmentation and frequency analysis of webpage keywords, It also studies how to classify webpages by weighted Chinese webpage keywords, involving technical fields such as automatic webpage acquisition, Chinese webpage preprocessing, Chinese word segmentation and keyword frequency analysis, and fuzzy classification of Chinese webpages. Background technique [0002] With the rapid development of Internet technology and Web technology, the number of web pages on the Internet is constantly increasing. The increase of network information greatly facilitates people to obtain information, but the excessive amount of information also brings a lot of difficulties for people to ...

AI Technical Summary

Problems solved by technology

[0005] Second, Chinese webpages contain a lot of "noise". Many webpages are not standardized, and contain a lot of advertisements, annotations and other information.

[0007] Third, most of the current research on the classification of Chinese webpages focuses on the feature selection and vector representation of Chinese webpages, and the use of the KNN classification algorithm to classify webpages. This method has a high time cost and complexity for webpage classification.