arrowCIRSS Home arrow Events arrow E-Research Roundtable arrow Event Details

E-Research Roundtable - Extracting Knowledge Claims for Automatic Evidence Synthesis Using Semantic Technology

Wednesday, November 9, 2016
12:30pm - 2pm

109 IS

Event Details

Session leaders: Kenney (Jinlong) Guo, iSchool PhD Candidate
Systematic review, a form of evidence synthesis that critically appraises existing studies on the same topic and synthesizes study results, helps reduce the evidence gap. However, creating a systematic review is very time and resource consuming. And most importantly, conclusions from a systematic review may go out of date quickly. While technologies (including text mining and machine learning) have started to be applied to facilitate the manual process of creating a systematic review (e.g. article screening, data extraction, etc.), so far there is little work on automatically detecting signals to update a systematic review. This is a challenging task because a) interpreting the conclusion of a systematic review is challenging and b) detecting new studies that may possibly change the conclusion of a systematic review is also difficult. A promising approach is to make semantic representation of the claims made in both the systematic review and the primary studies it synthesizes. In this presentation, I will introduce a taxonomy to represent knowledge claims both in systematic reviews and primary studies with the goal of automatically updating the conclusion of a systematic review. I will report initial results on developing such a taxonomy and also describe possible machine learning models to automatically extract the claim information for each step of the process.

Bio: Kenney (Jinlong) Guo is a doctoral student at the School of Information Sciences. His research focuses on using text mining and semantic technology to synthesize evidence from scientific literature.