Language Inference in Clinical Text
Brown Bag Lecture by Dr. Chaitanya Shivade | 10/2/2018 11:00AM – 12PM | 7th Floor Conference Room, Bldg 38A
Natural language inference (NLI) or recognizing textual entailment (RTE) has been a popular task among researchers for the past several years. A large body of work has investigated this problem in the open domain with popular datasets like SNLI and Multi-NLI. However, attention to this problem from the clinical community has been limited. This talk will discuss work from an upcoming EMNLP paper: 1) A new openly available dataset (MedNLI) for this problem in the clinical domain, 2) Models and methods for MedNLI, and 3) Techniques to incorporate domain specific knowledge (from corpora and ontologies) into neural networks for NLI.
More at: http://arxiv.org/abs/1808.06752
Dr. Shivade is a Research Staff Member with IBM at the Almaden Research Center in San Jose, California. He is part of the Medical Imaging Solutions group with a focus on Natural Language Processing (NLP) and Machine Learning (ML) for clinical data. Before joining IBM in 2016, Chaitanya earned a Ph.D. at the Ohio State University, co-advised by Eric Fosler-Lussier and Albert Lai. He enjoys inter-disciplinary research, and has publications across venues in Computer Science, Biomedical Informatics, and Linguistics.