Twitter sentiment analysis of DKI Jakarta's gubernatorial election 2017 with predictive and descriptive approaches

The main purpose of this research is to analyze sentiment of DKI Jakarta's gubernatorial election 2017 in social media Twitter with predictive and descriptive approaches. The dataset were collected from Twitter by using each candidate's username as the search query. For the predictive approach, machine learning algorithms such as Multinomial Naive Bayes and Support Vector Machine are used to classify the dataset. As for the descriptive approach, time series graphs and wordclouds are used to get the deeper insights of the dataset and find the connection between Twitter's sentiment and the result of the election itself.