Sentiment analysis towards Jokowi`s government using twitter data with convolutional neural network method

Abstract

Abstract In this current digital era, internet has become commonly uses in the society, it become an important aspect in life, one of its function are as a communication platform and channel for public opinion. In which utilize social media platforms. Indonesia really active in communication and conveyor expression using social media. In such a way that indonesia’s government takes advantage of it as a way to communicate and get feedback from indonesian people. In this research, as jokowi lead his second periode as a president of indonesian’s government, we try to see how indonesian sentiment toward his leadership. Using Convolutional Neural Network method from 7228 row data crawled from twitter obtained 38% positive sentiment, 51% negative sentiment, and 18% neutral sentiment. Overall we have 57% prediction accuracy using CNN model.