Finding all the Needles in the Haystack. A System to Estimate the Costs of e-Discovery and Systematic Reviews.
Giorgio Maria Di Nunzio · Padua Research Archive (University of Padova) · 2018
Systematic review and e-Discovery have a common task in which the objective is to find most (if not all) of the relevant documents in a collection by means of a (semi-)manual screening of the potentially interesting documents [3].However, the high cost of e-Discovery software and the management of the advanced e-Discovery mechanism are expected to affect the growth of this market (which is expected to reach 17.32 billion dollars by 2023). 1 Moreover, the large and growing number of published studies makes the task of identifying relevant studies in systematic reviews in an unbiased way both complex and time consuming [4].In this paper, we present an active learning system which combines different sampling approaches in order to estimate a 95% confidence interval of the number of relevant documents while taking into account the monetary costs of running the system itself.