Join us for a special presentation by Yue Wang, PhD candidate in the Department of Electrical Engineering and Computer Science at the University of Michigan, titled "Dual Intelligence: Interactive Machine Learning for Health Informatics" on Tuesday, Jan. 16, at 3:30 pm in Manning 01.
Bio: Yue Wang is a PhD candidate in the Department of Electrical Engineering and Computer Science at the University of Michigan. He is broadly interested in text data mining and machine learning with applications in health informatics. His thesis focuses on developing principled interactive machine learning approaches that reduce human analysts' information processing workload. His work is motivated by and applied to various data mining problems, including high-recall information retrieval, clinical natural language processing, and qualitative content analysis. He publishes in prestigious venues in both computer science and health informatics, including KDD, SIGIR, WSDM, AMIA and JAMIA. He serves as a program committee member for WSDM, SIGIR, CIKM, and WWW. He and his collaborators won the first place in TREC 2013 Microblog Track. He received the Best Paper Award and Outstanding Reviewer Award in WSDM 2016.
Abstract: Recent years witness an unprecedented growth of health data, including millions and millions of biomedical literature, electronic health records, and health forum posts. Information retrieval and data mining techniques are powerful tools to unlock the potential knowledge from these data, yet they need to be guided by human experts. Unlike training machine learning models in other domains, labeling health data requires highly specialized expertise, and the time of medical experts is extremely limited. How can we mine big health data with little expert effort? In this talk, I will present state-of-the-art interactive machine learning algorithms that bring together human intelligence and machine intelligence in health data mining tasks, using systematic review and clinical natural language processing as example use cases. By making efficient use of human expert's domain knowledge, we can achieve high-quality solutions with minimal effort. I will introduce future research plans along this direction.