<![CDATA[Twitter is one of the most widely used social media today. Besides being used as a social media, Twitter is also used to read news. Every year Twitter users have increased, so that information is also increasing. Increased information causes users who want to look for a certain information to experience difficulties. To solve the problem, news categorization is required. This study use Improved Naive Bayes method to categorize tweets by news contents. In Improved Naive Bayes posterior value will be calculated after the word is done by weighting using Bernoulli representation or by 1 and 0. This study use eight categories of news in Indonesia, which are: economy, entertainment, sports, technology, health, food, automotive, and travel. Based on the results of tests that have been done this study obtain precision value of 0.962961, recall 0.789164 and f-measure of 0.862973.]]>
Copyrights © 2017
