The spread of hoaxes in Indonesia has become a big concern for the public, especially now that the COVID-19 virus pandemic is hitting the whole world. Due to the large number of people who believe the hoax news regarding the COVID-19 vaccination that has spread on social media, many people refuse to carry out the COVID-19 vaccination as a form of government effort in dealing with this pandemic. Therefore, people need to be wiser when reading news on social networks. To help the public not to read hoaxes, it is necessary to classify the COVID-19 vaccine hoax. This study builds a system to classify hoax news on the COVID-19 vaccine. The model was built using the ensemble method by comparing the Random Forest and AdaBoost algorithms to choose a good classification for detecting hoaxes. In this research, there are use two test scenarios. The first scenario is an experiment using the Random Forest algorithm method and the second scenario is an experiment using the Adoboost algorithm method. The experimental results show that the first scenario produces a good accuracy value with the random forest algorithm method of 93.58%.