Demographic Classification is a method to classify people by its demographic. Twitter has become one of largest social media which there are millions of tweets posted every day. Indonesia also becomes one of major country who uses Twitter. In Twitter there is no way to know how to classify each user into its demographic attributes. Because in Twitter profile there’s no demographic attributes like gender or age. This research will focus on how to classify Indonesian Twitter users based on their gender, age and occupation. This research will use Naïve Bayes and K-Nearest Neighbor as its classifier algorithm. According to the result, Naïve Bayes performs well only in gender classification while K-Nearest Neighbor does not perform well in any demographic classification. Testing set successfully classified gender but failed with age and occupation