Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

SimpleLSTM: A Deep-Learning Approach to Simple-Claims Classification

: Chawla, P.; Esteves, D.; Pujar, K.; Lehmann, J.


Oliveira, P.M.:
Progress in Artificial Intelligence. 19th EPIA Conference on Artificial Intelligence, EPIA 2019. Proceedings. Pt.II : Vila Real, Portugal, September 3-6, 2019
Cham: Springer Nature, 2019 (Lecture Notes in Artificial Intelligence 11805)
ISBN: 978-3-030-30243-6 (Print)
ISBN: 978-3-030-30244-3 (Online)
European Conference in the Field of Artificial Intelligence (AI) <19, 2019, Vila Real>
Fraunhofer IAIS ()

The information on the internet suffers from noise and corrupt knowledge that may arise due to human and mechanical errors. To further exacerbate this problem, an ever-increasing amount of fake news on social media, or internet in general, has created another challenge to drawing correct information from the web. This huge sea of data makes it difficult for human fact checkers and journalists to assess all the information manually. In recent years Automated Fact-Checking has emerged as a branch of natural language processing devoted to achieving this feat. In this work, we give an overview of recent approaches, emphasizing on the key challenges faced during the development of such frameworks. We test existing solutions to perform claim classification on simple-claims and introduce a new model dubbed SimpleLSTM, which outperforms baselines by 11%, 10.2% and 18.7% on FEVER-Support, FEVER-Reject and 3-Class datasets respectively. The data, metadata and code are released as open-source and will be available at