Improving ranking for systematic reviews using query adaptation

Identifying relevant studies for inclusion in systematic reviews requires significant effort from human experts who manually screen large numbers of studies. The problem is made more difficult by the growing volume of medical literature and Information Retrieval techniques have proved to be useful to reduce workload. Reviewers are often interested in particular types of evidence such as Diagnostic Test Accuracy studies. This paper explores the use of query adaption to identify particular types of evidence and thereby reduce the workload placed on reviewers. A simple retrieval system that ranks studies using TF.IDF weighted cosine similarity was implemented. The Log-Likelihood, ChiSquared and Odds-Ratio lexical statistics and relevance feedback were used to generate sets of terms that indicate evidence relevant to Diagnostic Test Accuracy reviews. Experiments using a set of 80 systematic reviews from the CLEF2017 and CLEF2018 eHealth tasks demonstrate that the approach improves retrieval performance.