<![CDATA[The COVID-19 pandemic has hit Indonesia, many community activities are carried out at home. At that time, people often expressed their concerns through social media. One of the popular social media is Twitter which has a Tweets feature. Indonesian people who use Twitter use Tweets to write various opinions on the situation caused by COVID-19, be it government policies, vaccines, new variants of COVID-19 and so on. The diversity of these Tweets can be reflected in a section or group based on the context of the Tweets. The results of grouping Tweets can get opinions that are often expressed by the public about COVID-19. Where the results of the analysis can be used as reference material for the government in making policies during the COVID-19 pandemic. In this grouping using the BM25 method as a weighting and measuring Tweets. And K-Means Clustering where this algorithm is used. The results of the analysis and testing show that the number of terms must be reduced because the number of terms is a description of the many features used. a major feature that causes the BM25 method to be unable to distinguish the data. With the number of terms 20, parameters BM25 k1 = 1.2 and b = 0.5 and with a value of K = 3 will get the highest Silhouette Coefficient value, which is 0.3003]]>
Copyrights © 2022
