Question Answering (QA) paradigm can help to retrieve concise information in a natural way, given the precise answer and the supporting passages for any information needed. QA is widely applied in dialog systems, chatbots, designed and information retrieval in general. As QA is a classic field, most of the approaches are NLP linguistically-based. However, with recent developments in deep learning, neural network models have shown promise.
In this session, we will present a Deep Learning approach for Question Answering over biomedical data. Medicine is an evidence-based science, physicians most support their decision in Biomedical literature. In this field the development of methods that contribute to bypass the manual checking of candidate documents, is playing an important role. Along the session we will go through QA pipeline, starting from information extraction, representation, enrichment, pattern identification, model selection and finally answer extraction.
6:30pm-7:00pm - Networking/Intros
7:00pm-7:45pm - Talk by Andres Rosso (+QA)
7:45pm-8:30pm - Discussion/Networking
TheVentureCity has a rear lot with free parking, accessible via 9th St. Otherwise, you may use metered parking on 8th St.
About the speaker:
Andres is a Ph.D. Candidate at Universidad Nacional de Colombia in Bogotá. His research work is focused on deep learning and machine learning methods for Automatic Question Answering. Currently, he's working as a Data Scientist manager for Mo Technologies, leading the team responsible for developing machine learning models to asses the financial risk using traditional and external information sources.
His interest areas are: