Fake News and Hostile Posts Detection Using an Ensemble Learning Model

This paper describes the system submitted to Constraint 2021. The purpose of this task is to identify fake news in English and hostile posts in Hindi. We experimented with the pre-trained model based on the transformer and adopted the method of Ensemble Learning. We observed that the model ensemble was able to obtain better text classification results than a single model, the weighted fine-grained F1 score of our model in subtask B was 0.643998 (ranking 1/45).