Classification of Twitter User Sentiments Against Government Policies in Overcoming Covid-19 in Indonesia

Putranto, Hermawan Arief and Rizaldi, Taufiq and Dewanto, Wahyu Kurnia and Zahro, Rokhimatus (2022) Classification of Twitter User Sentiments Against Government Policies in Overcoming Covid-19 in Indonesia. Compiler, 11 (2). ISSN 2252-3839

	Text Classification of Twitter User Sentiment Against Government Policy in Overcoming Covid-19 in Indonesia.pdf Download (834kB)
	Text Similarity Classification of Twitter.pdf Download (1MB)

Abstract

Sentiment classification is a field of study that analyzes a person's opinions, sentiments, judgments, evaluations, attitudes, and emotions regarding a particular topic, service, product, individual, organization, or activity. The topic that is currently being discussed is Covid-19. Covid19 is a disease caused by a coronavirus, first identified in Wuhan City, China. This disease has spread throughout the world, one of which is Indonesia. Related to this, the Government of Indonesia issued a policy in an effort to break the chain of the spread of the coronavirus. However, this prompted the emergence of various kinds of community responses. One of them is Twitter users, there are pros and cons responses from the community in addressing government policies and causing problems, namely the difficulty of knowing positive, neutral, or negative responses given by the public. Based on the explanation above, sentiment analysis is carried out. This analysis was carried out by utilizing data from Twitter with the keywords dirumahaja, vaksinuntukrakyatindonesia, psbb, covid, covid19, covidindonesia, vaksinjakarta, vaksin, vaksinPulihkanRI, and vaksinDemiLindungiNKRI. Where the data will be processed through several stages, namely preprocessing, word weighting, and sentiment analysis. The results of the sentiment classification of the majority of Twitter users' responses are neutral, which are 69.2% of the data is classified as neutral sentiment, 30.1% of the data is classified as positive sentiment, and 7% of the data has negative sentiment

Actions (login required)

View Item