ANALISIS SENTIMEN PADA MEDIA SOSIAL TWITTER TERHADAP KEPOLISIAN MENGGUNAKAN ALGORITMA SUPPORT VECTOR MACHINE

Abstract

Sentiment analysis is one of the sciences in Text Mining which focuses on the classification of text documents that contain opinions or thoughts from the public. This research aims to gain an understanding of the public's views on the police through tweets posted on the social media Twitter. This research uses the Support Vector Machine algorithm. The text document used uses two labels, namely: positive and negative. In this research, 303 data were used and testing was carried out using the Confusion Matrix. The results of this study show that negative review polarization has a higher dominance compared to positive polarization, namely negative reviews are 51.16% while positive reviews are 48.84%. The results of the Confusion Matrix testing that was carried out obtained an accuracy rate of 54.10%, negative precision 56.41%, positive precision 50%, negative recall 66.67% and positive recall 39.29%.