Research On The Technology Of Web Information Resources Evaluation And Classification

Along with the rapid growth of the web resource, the web information becomes richer and richer. The quality of web information resource classification is directly determines how the current resource can serve efficiently information demanders.Web information resource quality will soon become the hot topic in the era of the proliferation of network resource. In order to enhance the predictability of information resource, this paper proposes a compensated stabilizer quality evaluation methods of resource, analyses, evaluates and separates the numerous and jumbled of resource, and merges the resource quality feature with the feature extraction method and the classification algorithm, reduces the pointless overhead of various matching, and extremely improves the efficiency and quality in the process of preprocess and categorization.This page brings forward a special industry oriented classification method with liability, which is to upgrade efficiency and practicality of industry web information resource classification. It makes semantic convertion to industry nomenclature by dint of industry knowledge storage, to relating to industry theme character. Then it picks up theme character of swatch, and uses evaluate functions, to calculate out weight quotiety. The calculate method is improved by combining character distributing complexion. The TFIDF weight quotiety calculate method of VSM is used to establish sorting rule, to realize the determinant of industry information resource. This page also combines KNN and experimented, to validate availability of the method.Finally, this page, aims at improving the mechanism of resource classification and increasing the attainability and accuracy of resource, etc, constructs an efficient and sound system of web e-government information classification, reflects, categorizes and stores dispersed resource based on the multi-index catalogue.