Claim Missing Document
Check
Articles

Found 1 Documents
Search
Journal : JURNAL ILMIAH INFORMATIKA

PERBANDINGAN ALGORITMA NAÏVE BAYES CLASSIFIER DAN K-NEAREST NEIGHBORS UNTUK ANALISIS SENTIMEN COVID-19 DI TWITTER Habibi Aulia Nur Syifa Habibi; Arie Nugroho; Rina Firliana
JURNAL ILMIAH INFORMATIKA Vol 11 No 01 (2023): Jurnal Ilmiah Informatika (JIF)
Publisher : LPPM Universitas Putera Batam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33884/jif.v11i01.7069

Abstract

COVID-19 emerged in China in 2019. In Indonesia in 2020 there were more than 3000 positive cases of COVID-19 with a mortality rate of 9.1%. The government's incomplete efforts to break the chain of the spread of COVID-19 have made people uneasy about this pandemic. Many people want to express their aspirations on social media which is considered suitable as a place that represents the aspirations of the COVID-19 pandemic. One of them is twitter. There are so many text messages sent, some positive and some negative, that it is difficult to retrieve harmonized information due to the diversity of text messages sent. One way to overcome this is with sentiment analysis. This research has processes including text preprocessing, word weighting, classification with K-Nearest Neighbors (KNN) and Naïve Bayes Classifiers (NBC) algorithms. The results obtained by KNN got an accuracy of 72.37% while NBC amounted to 67.84%. KNN is the best classification algorithm for negative sentiment classification, the negative label predicted correctly in KNN is greater, namely 393 compared to NBC which is 339. while NBC is the best algorithm for positive sentiment classification, the positive label predicted correctly NBC is 275 greater than KNN as much as 262.