Classification of Breaking News Taken from the Online News Sites

Abstract

In this study, we aimed to provide access to the breaking news depending on the category to which the user wants. First, accessing to news in certain categories are provided from the news provider by using RSS (Really Simple Syndication). Preprocessing is implemented by cleaning xml tags and punctuation which can cause illusions before the content are obtained on datum. The features which can represent our classes in categories were determined by applying the methods in data mining for content after preprocessing phase. In the last step of process, Classification of category process is done by obtaining breaking news' content taken as online. In the phase of classification, Categorization were implemented with features which represent each category and by using C4.5i Naive Bayes and SMO (Sequential minimal optimization) functions, respectively. The performance rates in the usage methods and classification rates are shown in comparison.