Currently, the training dataset has only 2756 tweets, which is a very low amount of data for training. * Add as much data as possible, but also see that the data is correct.
Currently, the training dataset has only 2756 tweets, which is a very low amount of data for training.