Uji Akurasi Penggunaan Metode KNN dalam Analisis Sentimen Kenaikan Harga BBM pada Media Twitter

Abstract

The government in an effort to run the government is inseparable from the policies that must be made and taken. One policy alone can attract diverse sentiments from society. Based on this, this study was made to analyze public sentiment towards government policies, especially government policies regarding fuel price hikes. In this study, the analysis process using the K-Nearest Neighbor algorithm classifies tweets from Twitter into two categories, namely positive and negative. The research stages started from crawling data, data preprocessing, labeling, classification using the KNN algorithm, and evaluation. With an accuracy of 94.33% in classifying data. With the results of this research, it is hoped that it will make it easier for the government to see people's responses and sentiments towards the fuel increase policy so that the government can produce better policies by incorporating what the people have given.