SMATIKA
Vol 11 No 01 (2021): SMATIKA Jurnal : STIKI Informatika Jurnal

Analisis Sentimen Masyarakat di Twitter Terkait Pandemi Covid-19

Yessi Yunitasari (Unknown)
Andi Rahman Putera (Unknown)



Article Info

Publish Date
22 Jun 2021

Abstract

Social media is a place to look for new friends and a place to express opinions about something freely. One social media that is widely used today is Twitter. Many people use Twitter to issue opinions on the Covid-19 Pandemic that occurred in various countries, including in Indonesia. The Covid-19 pandemic or coronavirus in Indonesia begins with the discovery of a 2019 coronavirus sufferer (COVID-19) on March 2, 2020, to April 8, 2020, 2,738 positive cases of COVID-19 have been confirmed, with 221 cases of which have died and 204 cases have recovered. Tweets written by the public can later be classified into positive and negative sentiments using sentiment analysis. The results of our sentiment analysis can see how the perceptions of the Indonesian people regarding the Covid-19 pandemic that occurred in Indonesia. The sentiment analysis classification process will use the naïve Bayes method. Sentiment testing carried out using Cross-Validation includes 5 Fold and 10 Fold testing. From each of these tests, the accuracy, precision, and recall values will be seen. the results of the Cross Validation 5 Fold test obtained results from an average accuracy of 0.756364 (75%). The results of the Cross Validation test for 10 Fold obtained results from an average accuracy of 0.76 (76%)

Copyrights © 2021