题目：Large Scale Medical Natural Language Processing

主讲人：郝天永  广东外语外贸大学信息学院 副教授/博士

时间：5月6日(周五)下午2:10-3:40

地点:公司301室

欢迎感兴趣的师生参加!

 Abstract:

A large amount of medical texts generate highly relevant evidences for effective disease treatments. However, extracting needed information from such big data through natural language processing remains a research problem due to the complex of medical texts. We collaborated with Columbia University based on EliXR project and made some progress on large scale clinical trial text processing. This talk will briefly present our research on: 1) extensible automated semantic tag mining; 2) TEXer - temporal expression extracting and normalizing; 3) TrialSmart - towards personalized search of clinical trials through preference learning; 4) clinical trial clustering by similar semantic phenotypes. In particular, we will introduce Valx – an automated approach for numeric expression extraction and normalization and its application on medical text processing. We finally demonstrate some of our developed NLP systems.

  Bio:

Dr. Tianyong Hao is an associate professor in Guangdong University of Foreign Studies. He received Ph.D. in Computer Science from CityU of HK in 2010. He visited York University (2008), Emory University (2009-2010), University of New South Wales (2012), and then worked as Postdoc research scientist at Columbia University in New York City (2013-2014). His research interests include Question Answering, Medical Informatics, and Computational Linguistics.

Dr. Hao is IEEE Senior member and chair/co-chair/session-chair for more than 10 international conferences. He has published more than 50 refereed papers, in which 11 are ISI-JCR Q1/Q2 SCI journal papers and 35 are EI indexed. He is the PI(co-PI) of several grants including NSFC and UGC GRF, as well as an applicant of 6 patents. He received best paper/best student paper award from 3 conferences including 10th IEEE ICCI’CC(Canada) and 6th WEBIST(Spain).