Background: machine learning and citizen science initiatives within Cochrane are transforming Cochrane’s centralised efforts to identify reports of trials. The RCT Classifier, a machine learning routine, and Cochrane Crowd are highly accurate and efficient at distinguishing between records that are likely to describe randomized controlled trials (RCTs) and those that are not. ‘Screen4Me’ is a new service that enables individual Cochrane Reviews to take advantage of the efficiencies that this combination of human and machine effort offers in individual reviews.
Specifically the new workflow makes use of three components:
- existing data – no more rescreening of results that have already been screened and deemed to be either an RCT or not;
- machine learning in the form of the RCT classifiers for both bibliographic and trial registry records;
- crowdsourcing via Cochrane’s citizen science platform Cochrane Crowd (crowd.cochrane.org).
Objectives: participants will:
- learn what Screen4Me is and when Screen4Me is appropriate to use;
- gain a good understanding of the Screen4Me workflow;
- try out the Screen4Me for themselves on a training dataset;
- discuss the range of possible use cases for this new workflow;
- feed in to further development of the individual tools and the workflow itself.
Description: this workshop is primarily aimed at Cochrane Information Specialists and Cochrane Review author teams. It will begin with a brief presentation of the Screen4Me workflow and the work conducted to date. The main part of the workshop will be hands-on. Participants working individually or in small groups will be supplied with a practice dataset representing the search results for a Cochrane Review. They will have the opportunity to send these results through the Screen4Me workflow via the Cochrane Register of Studies (CRS). Within the workflow, they will see the output from the RCT Classifier, as well as set up a task for Cochrane Crowd. The final 20 minutes of the workshop will be a question-and-answer discussion session.