RobotAnalyst: An online system to support citation screening in evidence reviewing

Article type
Authors
Nolan K1, Ananiadou S2, Przybyła P3, Brockmeier AJ4, Mcleod C1
1NICE
2School of Computer Science, University of Manchester
3National Centre for Text Mining, University of Manchester
4University of Manchester
Abstract
Objectives:
•Provide an overview of text-mining methods;
•Provide an interactive demonstration and training of the freely available RobotAnalyst software;
•Enable participants to be confident in using RobotAnalyst for screening search results for systematic reviews.

Description: The National Centre for Text Mining (NaCTeM) in collaboration with National Institute for Health and Care Excellence (NICE) have developed RobotAnalyst, an online programme for text mining and sifting databases to support the development of systematic reviews. The system incorporates machine-learning functionality (text mining) with visual representation of literature search results to enable efficient sifting of evidence for systematic reviewing. Initial evaluation of the system has shown promising results and efficiencies in time spent sifting 100% of a database. This workshop will allow delegates the opportunity to learn how to use the tool and explore its functionality.

The first part will provide an overview of the RobotAnalyst system. Participants will be walked through its key functionality including semi-automatic citation searching, faceted and topic-based searches.
The second part will allow participants to have hands on experience of using RobotAnalyst. Participants will work through a series of tasks to equip them with the skills needed to use the system.

Participants will need to bring a laptop to access the system.