Analisis Sentimen Kurikulum 2013 Pada Sosial Media Twitter Menggunakan Metode K-Nearest Neighbor dan Feature Selection Query Expansion Ranking

Nurul Dyah Mentari (Fakultas Ilmu Komputer, Universitas Brawijaya)
Mochammad Ali Fauzi (Fakultas Ilmu Komputer, Universitas Brawijaya)
Lailil Muflikhah (Fakultas Ilmu Komputer, Universitas Brawijaya)

Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer
Vol 2 No 8 (2018): Agustus 2018

Kurikulum 2013 has become a hot topic that is often discussed by society on Twitter. Twitter is one of the social media that used by a society to talk about a particular subject. This study attempted to analyze tweets about the Kurikulum 2013 by classifying whether it is a positive opinion or a negative opinion. Classification process is done by K-Nearest Neighbor method by using Query Expansion Ranking method for feature selection. There are 4 main processes in this analysis sentiment system that first is text pre-processing, term weighting (TF-IDF), feature selection, and classification. Based on the tests in this study proven that feature selection improve accuracy of systemresults. The best acuracy results of 96.36%was obtained when k = 1 and using a feature selection of 50% ratio. The test results by using selection feature of 50% ratio get higher accuracy than a system does not use the selection feature because some noise features that have been removed.