Special issue on technology assisted review systems

Abstract

When faced with an information need, an exhaustive search aim to retrieve all the relevant information. The need for comprehensive research is fundamental electronic discovery, systematic reviews, investigation, research in general, and also in the creation of evaluation collections in information retrieval. To support these high-recall retrieval tasks, technology relies on machine learning to develop systems that rank and classify documents with greater efficiency than traditional Boolean keyword searches. Technology-Assisted Review (TAR) is one of the key technologies which can help to improve this process. The aim of this special issue is to investigate the most recent approaches of TAR systems applied in a wide variety of contexts where high recall is necessary: from the classification of emails to the compilation of systematic reviews up to the evaluation of the quality of the assessments.