Dr. Shabnam Tafreshi's research interests are in computational semantic understanding across languages. Dr. Trafreshi is working working towards improving NLP models to better extract knowledge from the text by complementing NLP models with meaning constructs. Better models are needed for downstream tasks in text classification and knowledge extraction. In addition, her research is focused on methodologies for collecting and annotating quality data for building robust NLP models.
Confirm or Refuse: A Comparative Study on Citation Sentiment Classification in Clinical Research Publications, H Kilicoglu, Z Penga, S Tafreshi, T Tran, G Rosemblat, J Schneider, Journal of Biomedical Informatics 2019
Rhetorical Moves and Audience Considerations in the Discussion Sections of Randomized Controlled Trials of Health Interventions, J Schneider, G Rosemblat, S Tafreshi, H Kilicoglu, 2nd European Conference on Argumentation: Argumentation and Inference 2017