<![CDATA[Twitter social media is one way to get fast information, especially related to face-to-face learning system where during covid-19 pandemic learning is held online. In this case government has informed related to the face-to-face learning system as well as the community or students gave an enthusiastic response to the policies provided by the government including giving a good response to these policies and some of them disagreeing with these policies. In this case, the researcher analyzes public opinion on government policies related to face-to-face learning on Twitter social media using the Support Vector Machine algorithm. By doing an analysis related to government policies regarding learning during the COVID-19 pandemic, the government can find out how the public responds and can make decisions. Based on a series of processes that have been carried out previously using the Support Vector Machine method by applying the TF-IDF weighting function, the results can reach 93%. To see the level of accuracy of the proposed method, the researchers made a comparison by applying several other methods. The accuracy results obtained from the support vector machine method are 93%, based on the accuracy obtained, it can be determined that the level of accuracy using the Support Vector Machine method is quite high in classifying sentiment data, but when compared to other methods, namely nave Bayes, which obtains an accuracy of 94%, Logistic Regression which obtained 93% accuracy, and K-NN which obtained 90% accuracy. Thus, the accuracy results of four methods are not too different.]]>
Copyrights © 2023
