This is a front-end for our machine learning model that identifies reports of randomized controlled articles (RCTs). We have extensively validated the recall and precision of this model, as described in our Research Synthesis Methods publication, Machine learning for identifying Randomized Controlled Trials: An evaluation and practitioner's guide.
We are enormously grateful to the Cochrane Crowd project (both the team, and the volunteers), who have generously allowed us to use their data. Their hundreds of thousands of manually labelled abstracts allowed us to build this system.
Please cite our paper if you use the tool in published work:
IJ Marshall, A Noel‐Storr, J Kuiper, J Thomas, BC Wallace. Machine learning for identifying Randomized Controlled Trials: An evaluation and practitioner's guide. Res Syn Meth. 2018;1–12. https://doi.org/10.1002/jrsm.1287
This work is supported by: the NIH/NLM grant R01-LM012086-01A1, NIH/NCI grant 5UH2CA203711-02, and also by the UK Medical Research Council (MRC), through its Skills Development Fellowship program, grant MR/N015185/1. We make use of data generously provided by Cochrane Crowd.