Feature Extraction of Tweet data Characteristics to Determine Community Habits

Social media is used as a place of interaction between users. On Twitter social media there is a lot of data that can be used, especially community data that will be used to explore the characteristics of the data. In this study, there are several characteristics of community data such as tweet data, user data, slang words, standard words and non-standard words. all of the data is used to determine user habits, this study creates a model that can determine user habits. This research uses a regression tree whose process will check the frequency of the most words for each user in the community tweet, then check the highest frequency and looping to get a pattern, the test carried out by the model from the Regression Tree can determine habits with test data as many as 4 users then it is found that the habit in the Indonesian standup community is “Tonton”.