K-Nearest Neighbor (KNN) untuk Menganalisis Sentimen terhadap Kebijakan Merdeka Belajar Kampus Merdeka pada Komentar Twitter

  • Rahmad Gunawan
  • Rahman Septiadi
  • Febby Apri Wenando
  • Harun Mukhtar Universitas Muhammadiyah Riau
  • Syahril

Abstract

On December 11, 2019, the Minister of Education and Culture of the Republic of Indonesia Nadiem Anwar Makarim issued a policy of "Merdeka Belajar". Netizens on Twitter have debated this Merdeka Belajar and became a trending topic. This study tries to analyze the sentiment of tweets about opinions on this policy by classifying whether it is a positive opinion or a negative opinion. The classification method applied is the K-Nearest Neighbor algorithm. In this study, four main processes were carried out, namely text-preprocessing, word-weighting (TF-IDF), classification and validation using k-fold cross validation. Tests were carried out with a dataset of 700 data, training was carried out using 630 training data and 70 testing data. In testing, the highest accuracy of the K-Nearest Neighbor algorithm was obtained at the k-8 value, namely 84.28%. Furthermore, validation is carried out using k-fold cross validation with a value of fold = 10 to get an accuracy of 84.42%.

Downloads

Download data is not yet available.
Published
2022-08-18
Abstract views: 83 , pdf (Bahasa Indonesia) downloads: 62