arrowCIRSS Home arrow Events arrow CIRSS Seminars arrow Event Details

CIRSS Seminar - Can Natural Language Processing Help in Fostering Responsible Research Practices?

Friday, November 11, 2016
4pm - 5pm

Room 126 IS

Event Details

Session leaders: Halil Kilicoglu, NLM/NIH staff scientist
Description: Billions of dollars are spent on biomedical research in U.S. every year. While this has led to significant advances in medicine and healthcare, research rigor and integrity issues continue to plague biomedical research, leading to what is sometimes termed "reproducibility crisis". The responsibility of research integrity ultimately lies with the researchers; however, journals, academic institutions, and funding agencies also have a responsibility in addressing the issue. As research output increases exponentially, a variety of practical tools are needed to assist these stakeholders in doing their part properly and in a scalable manner.
Textual artifacts (e.g., grant proposals, manuscript submissions, publications) are central to the biomedical research enterprise. With the progress made in in natural language processing (NLP) and biomedical text mining (bioNLP) in recent years, it is timely to ask whether the advances in these fields can inform development of tools and systems, which can contribute to improving research integrity and fostering responsible practices. While there has been little NLP research specifically focusing on research integrity so far, methods developed for tasks such as text classification, information extraction, automatic summarization, and rhetorical analysis can play a significant role in addressing the challenge. In this talk, I will motivate the use of NLP to ensure integrity of research and identify several core tasks towards this goal (e.g., plagiarism detection, claim tracking/validation, citation distortion detection). I will discuss the NLP techniques that can be employed to tackle such research integrity tasks, how mature these techniques are and how they can be adapted and enhanced. I will also highlight the challenges in adoption of such technology for practical use.

Bio: Dr. Halil Kilicoglu is a staff scientist at the National Library of Medicine, National Institutes of Health. He earned his PhD in Computer Science from Concordia University, his MS in Computer Science from George Washington University, and his BS in Computer Engineering from Istanbul Technical University. Dr. Kilicoglu has expertise in biomedical natural language processing and text mining; more specifically, his research focuses on semantic interpretation of biomedical text, question answering, and literature-based knowledge discovery.

[Hosted by Jodi Schneider.]