Abstract Automatic classification of text is one of the important applications and search subject ago the Foundation of digital document. Text classification is necessary and the reason excessive number of document text handled daily. This study mainly aims to integrate data mining techniques with classification tasks to build a web text classification (TC) method by 1) influencing the full amount of information contained within web documents for classification and 2) increasing the effectiveness of search processes to identify similar or related information for users. Hypertext classification differs from traditional TC; information beyond web document contents, such as metadata, is a useful source of information. The unification of metadata with text using simple combination methods effectively improves classification performance. Results show that these classifiers are common, accurate, and perform effectively. The accuracy measure is 91%.
Alan : Mühendislik
Dergi Türü : Uluslararası
Benzer Makaleler | Yazar | # |
---|
Makale | Yazar | # |
---|