Files in this item

FilesDescriptionFormat

application/pdf

application/pdfAnalyzingInclusion_JingyiXie.pdf (4MB)
(no description provided)PDF

Description

Title:Analyzing Inclusion Criteria of 7000 Cochrane Systematic Reviews
Author(s):Dong, Xiaoru; Xie, Jingyi
Contributor(s):Schneider, Jodi; Hoang, Linh
Subject(s):Systematic Review
Classification
Classifier
Machine Learning
Abstract:In healthcare, a systematic review is a type of literature review designed to synthesize all available evidence on a given question. To do so, systematic reviewers start with inclusion criteria that help to identify relevant studies that could be included into the review. Identifying what evidence should be included is critical in the systematic reviewing process because it helps reviewers crisply identify what they are looking for, which saves time and effort to screen through the literature. In this project, our objective is to build a machine classifier that categorizes textual descriptions of inclusion criteria taken from over 7000 Cochrane Systematic Reviews to determine whether or not the included articles must be Randomized Control Trials. The ultimate goal of developing such a tool is to analyze the inclusion criteria of existing systematic reviews, to understand what types of evidence are considered relevant for systematic reviews, and to identify Randomized Control Trials in included studies. Linh Hoang is funded by National Library of Medicine: "Text Mining Pipeline to Accelerate Systematic Reviews in Evidence-based Medicine" (R01LM010817). Thank you to Cochrane for providing Cochrane reviews as machine readable XML from which inclusion criteria were extracted. This research was done in collaboration with the Department of Statistics, College of Liberal Arts and Sciences and the School of Information Sciences at the University of Illinois at Urbana-Champaign.
Issue Date:2018
Genre:Conference Poster
Type:Text
URI:http://hdl.handle.net/2142/102246
Date Available in IDEALS:2019-01-11


This item appears in the following Collection(s)

Item Statistics