Automating Systematic Reviews

How machine learning could reduce the literature review process by months

EDINA have been working with SEBI (Support Evidence-Based Interventions) in the Royal Dick School of Veterinary Medicine on the automation of systematic reviews. Performing literature reviews for the Gates Foundation, SEBI found that using machine learning techniques to do data mining on the literature databases could turn a three month manual effort into one taking only three weeks - further automation is seeking to turn that into three days!

As a trial the same approach was tried with the Engineering school, for the Microfluidics department. This showed good results, and demostrates that the technology is quite transferrable.

EDINA are interested in hearing from anyone who is involved in doing systematic reviews on a regular basis, with a view to examining the possibilities for automation. The approach is likely most effective for anyone performing literature reviews involving data mining from databases like Google Scholar and PubMed, but we are interested in speaking to anyone who might benefit from automation. Please contact Andrew Horne for further information.

Stay updated about further developments by subscribing to our mailing list.


Click Here to Subscribe