Analisis Sentimen Konten Radikal dalam Kontestasi Politik 2019 di Media Twitter Menggunakan Interjection dan Punctuation

Abstract

Sentiment analysis is now a trend to identify people's opinions and emotions in responding to a situation. In the political year, many opinions were scattered both written in print and social media. Political actors have different views, so that raises a lot of opinions that lead to radical actions such as SARA to people with different opinions. Research related to the analysis of radical sentiments via Twitter has been done by several researchers before, but there have been no studies of radical sentiment analysis using extraction features. This study proposes to conduct a radical content sentiment analysis on Indonesian textual tweets related to political contestation in Indonesia which then uses two features namely punctuation and interjection, and is classified using the support vector machine algorithm. From the results of the classification that has been done, obtained an accuracy value of 80% sentiment analysis and radical sentiment analysis conducted several times with a number of different interjection, obtained an accuracy of 94% using 200 interjection words.