ANALISIS SENTIMEN TERHADAP VAKSIN COVID-19 PADA TWITTER MENGGUNAKAN METODE LEARNING VECTOR QUANTIZATION

Indonesia is one of the countries affected by the COVID-19 outbreak, so the government has implemented various policies to break the chain of spread of COVID-19, one of which is by vaccinating. With the Covid-19 vaccination program, many community groups reject this activity, so to make it easier for related parties to carry out vaccination activities, it is necessary to see how the response from the community is, one of which is by analyzing sentiment on Twitter social media. In this study, a dataset of 300 tweets was used with the distribution of 100 neutral data, 100 positive data, and 100 negative data using the learning vector quantization (LVQ) method and a comparison will be made with the support vector machine (SVM) method, as for the distribution of data, 20% test data and 80% training data. Before conducting the test, the Twitter data crawling stage was carried out using the Twitter API, then preprocessing and calculating the weight of the document using TF-IDF first. Based on the results of the research, the LVQ method obtained 74% precision, 91% recall, 82% f-measure with 73% accuracy, no more than SVM method with 80% accuracy with 87% precision, 91% recall, f-measure 89%