Training and Prediction Data Discrepancies: Challenges of Text Classification with Noisy, Historical Data R Andrew Kreek author Emilia Apostolova author 2018-11 text Proceedings of the 2018 EMNLP Workshop W-NUT: The 4th Workshop on Noisy User-generated Text Wei Xu editor Alan Ritter editor Tim Baldwin editor Afshin Rahimi editor Association for Computational Linguistics Brussels, Belgium conference publication kreek-apostolova-2018-training 10.18653/v1/W18-6114 https://aclanthology.org/W18-6114/ 2018-11 104 109