Text Categorization Using Naive Bayes Algorithm

• 第一著者: Wan Hazimah, Wan Ismail
• フォーマット: 学位論文
• 言語: English; English
• 出版事項: 2005
• 主題: QA71-90 Instruments and machines
• オンライン･アクセス: http://etd.uum.edu.my/1368/1/WAN_HAZIMAH_BT._WAN_ISMAIL.pdf; http://etd.uum.edu.my/1368/2/1.WAN_HAZIMAH_BT._WAN_ISMAIL.pdf; http://etd.uum.edu.my/1368/

As the volume of information available on the internet and corporate intranet continues to increase, there is a growing interest in helping people better find, filter, and manage all these resources. Text categorization is one of the techniques that can be applied in this situation. This paper presents text categorization system based on naive Bayes algorithm. This algorithm has long been used for text categorization tasks. Naive Bayes classifier is based on probability model that integrate strong independence assumptions which often have no bearing in reality. The aims of this project are to categorize the textual document using naive Bayes algorithm and to measure the correctness of the chosen technique for the categorization process. This paper also discusses the experiment in categorizing articles using naive Bayes. The result shows that the accuracy for training is 81.82% whereas the accuracy for testing is 47.62%.