A small reminder: In supervised text classification, we train a statistical model on the features of our data (e.g. the word frequencies) to predict the class of our texts (e.g. the sentiment, topic,… ...