Automatic classification using concept knowledge of web documents

• Main Authors: Choi, Sang-Ho, Park, Sa-Joon, Hwang, Su-Cheol, Kim, Ki-Tae
• Format: Conference or Workshop Item
• Language: English
• Published: 2004
• Subjects: QA76 Computer software
• Online Access: http://repo.uum.edu.my/13843/1/KM112.pdf; http://repo.uum.edu.my/13843/; http://www.kmice.cms.net.my
• Institution: Universiti Utara Malaysia

In order to classify web documents, we suggest a method using concept knowledge of category.In our study, the concept relations between keywords are extracted using hyperlink information and after the extracted keywords are classified into each category, these are used as an index.Then TFIDF for each category is extended to determine index weight value.The system is constructed for experimenting and estimating,which is consist of web robot, indexer, concept knowledge database for each category and the document classifier.Our system to be applied the extended TFIDF method shows an accuracy of 88% in automatic classifying of web documents.