期刊名称:International Journal of Artificial Intelligence & Applications (IJAIA)
印刷版ISSN:0976-2191
电子版ISSN:0975-900X
出版年度:2017
卷号:8
期号:6
页码:1
出版社:Academy & Industry Research Collaboration Center (AIRCC)
摘要:With growing texts of electronic documents used in many applications, a fast and accurate textclassification method is very important. Arabic text classification is one of the most challenging topics. Thisis probably caused by the fact that Arabic words have unlimited variation in the meaning, in addition to theproblems that are specific to Arabic language only. Many studies have been proved that Naive Bayes (NB)classifier is being relatively robust, easy to implement, fast, and accurate for many different fields such astext classification. However, non-linear classification and strong violations of the independenceassumptions problems can lead to very poor performance of NB classifier. In this paper, first, we preprocessthe Arabic documents to tokenize only the Arabic words. Second, we convert those words intovectors using term frequency and inverse document frequency (TF-IDF) technique. Third, we propose anefficient approach based on Kernel Naive Bayes (KNB) classifier to solve the non-linearity problem ofArabic text classification. Finally, experimental results and performance evaluation on our collecteddataset of Arabic topic mining corpus are presented, showing the effectiveness of the proposed KNBclassifier against other baseline classifiers.