Associate Professor Halil Kilicoglu will lead a session on "Natural Language Processing in Support of Meta-Research Investigations: Two Case Studies."
Abstract: Meta-research (also called "science on science") is a relatively new discipline, aiming to understand and improve the ways in which scientific research is conducted, disseminated, assessed, and rewarded. Scientific publications represent a rich source of data for meta-research investigations. In this talk, I will first argue for the use of natural language processing (NLP) techniques in meta-research studies. Next, I will present two recent projects that apply NLP methods to study reporting in biomedical publications. The first study investigates the impact of peer review on discussion of study limitations and strength of claims in randomized controlled trial (RCT) publications. In the second study, we propose an information extraction approach to identify variables relevant to translatability of pre-clinical animal studies to human populations.
Kilicoglu's research is primarily concerned with natural language understanding, with a particular focus on biomedical text. He uses a combination of data-driven analytical techniques and knowledge-based semantic approaches to extract and organize knowledge buried in textual artifacts, with potential benefits for biomedical discovery and scholarship, and healthcare outcomes.
Questions? Contact Janet Eke
This event is sponsored by CIRSS