ABSTRACT
The spread of COVID-19 has recently become a public concern. There are many public emotions regarding implementing the Large-Scale Social Restrictions (PSBB), which was especially implemented in Jakarta, first implemented in Indonesia. People have various emotions mirroring their tweets in making statements on social media, especially Twitter. Emotional expressions can be joy, sadness, anger, and fear. This study aims to determine the impact of the implementation of PSBB in reducing the spread of COVID-19 on people's emotional factors on Twitter. The method used in this research is the SentiStrength method and Support Vector Machine. Furthermore, the comparison between the two methods is completed to determine which one is better. The tweet data used were 12,735 lines from 10 April 2020 to 21 August 2020. The highest accuracy achieved of SentiStrength and SVM is 88.33% and 73.33%, respectively. Similarly, f-measure of SentiStrength (88.14%) outperforms SVM (75%). This research shows that the implementation of PSBB on public emotional factors on Twitter is that happy emotions with the highest sentiment are positive sentiments, reaching 5246 sentiments. © 2022. All Rights Reserved.
ABSTRACT
To accelerate the handling of the spread of COVID-19 in Indonesia, the Government of the Republic of Indonesia has issued a vaccination discourse for the people of Indonesia;as of January 20, 2021, there are more than 939.000 incidents and more than 26.000 incidents resulting in mortality. With the quick transmission of COVID-19 and the dangers it caused, the Indonesian government has taken precautions by vaccinating the people. This vaccination information spread on various social media, including Twitter which has a comment feature on its posts;it cannot automatically verify the volume of user sentiment regarding positive or negative comments. Sentiment analysis does partially of what text mining does for text splitting in knowing the polarity of an idea using a particular algorithm. It is positive or negative polarity. This study aims to provide results of public opinion on sentiment analysis of COVID-19 Vaccination using the Lexicon Based method. This research found that 95.4% of Tweets or messages related to Vaccination on Twitter are messages with neutral sentiments, 1.95% Slightly Positive, 1.81% positive sentiments, 0.56% Slightly Negative, and 0.28% are negative sentiments. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.