Haber metinlerinin farkli metin madenciliği yöntemleriyle siniflandirilmasi

With the development of technology, people are entering the virtual world more and more. Parallel to this, the internet becomes a bigger network every day and it gets a complex structure depending on this growth. Achieving the desired information with structred data becomes an increasingly important problem. One of the useful ways to find solution for this problem is to divide this complex data into categories by text mining methods. By creating semantic similarities with this categorization, data can be achieved effectively and quickly. In this study, it is aimed to classify the news text data that have four different categories (economy, politics, sports and health) with different feature extraction and term weighting methods using different text mining techniques and to test the efficiency and success of the methods. By the proposed method, 100% classification success rate was obtained on news texts.