Home > Events > CLIP Colloquium: Han-Chin Shing (CS)

CLIP Colloquium: Han-Chin Shing (CS)

Time: 
Wednesday, March 04, 2020 - 11:00 AM to 12:00 PM
Location: 
5105 Iribe Center

 

A Prioritization Model for Suicidality Risk Assessment

Abstract: Suicidality is a problem whose importance cannot be overstated. Recently, using a classification framework, machine learning has shown promising results addressing the problem of suicidality. In this talk, I will argue that we should reframe suicide risk assessment from social media as a ranking problem whose goal is maximizing detection of severely at-risk individuals given the time available. Building on measures developed for resource-bounded document retrieval, I will introduce a well-founded evaluation paradigm, and demonstrate using an expert-annotated test collection that meaningful improvements over plausible cascade model baselines can be achieved using an approach that jointly ranks individuals and their social media posts.

Bio: Han-Chin Shing is a Computer Science Ph.D. student at the University of Maryland and a member of CLIP. His advisors are Philip Resnik and Doug Oard. He completed his M.S. in Computer Science at the University of Maryland, and his B.S. in Electrical Engineering at National Taiwan University. He works on natural language processing (NLP) and its application to the broader healthcare problem: automatic medical coding using clinical notes and suicide assessment using online postings. Currently, his focus is on using NLP to assist clinical experts to prioritize risk assessment of suicidality.