<![CDATA[Sentiment analysis is the interpretation and classification of user emotions (positive, negative, neutral) about a subject in text data using text analysis. Multilingual sentiment analysis is the process of assessing sentiment in more than one language. The tricky thing about being multilingual is that the emotions and behavior of our consumers are heavily influenced by culture and language. Therefore, for organizations with an international customer or user base, sentiment analysis is highly recommended to perform analysis not only in one language but in many languages. This is because the accuracy of the assessment will be better if it is done in more than one language. There are several methods that can be used to perform sentiment analysis, one of which is machine learning. Machine Learning is used as a tool to produce robots that are able to classify types of sentiment in textual data. This research was conducted to produce a machine learning model that can be used to measure the level of popularity of research objects based on comments written in two languages, namely Indonesian and English. The research was conducted based on comments on Twitter about the Lake Toba tourist attraction. Based on the results of testing the Naïve Bayes model on data testing, it shows that sentiment with positive predictions is 1,474 records or 54.03% and sentiment with negative predictions is 1,254 or 45.96%, with an accuracy rate of the method used at 97.1%.]]>
