Hi Teresa,

I don’t know if ‘toyed’ is the right word since I know of at least one person that has done their doctorate degree in computer sciences on this topic. It wasn’t particularly for critical appraisal, but rather for using computer-learning to assist in screening citations. Each year, the Cochrane Colloquium and other major KS conferences have presented the experience of teams using similar technology to assist in screening. I see this as the next major step forward that will bring KS into the digital age. We already use algorithms to search through millions of citations in bibliographic databases (removing the majority of the noise) and hopefully bringing back the most useful citations for a closer look by a ‘human’ operator. Even the ‘human’ operator has to be trained to identify key words, phrases, etc.

A great example of this type of technology that we all use every day (especially since this is a listserv) is spam filters. They usually use a naïve Bayesian approach, while more advanced ones maybe using computer-learning, to determine if the emails you receive are legitimate or ‘spam’. So you train it to recognize that words like Viagra, V I A G R A, V.I.A.G.R.A., Vairga (intentional misspelling), etc. in the subject header are most likely to be unwanted emails and should not make it to your inbox. I’m personally ‘toying’ around with the algorithms and would love to be able to use this in real-life.

Ahmed

From: Evidence based health (EBH) [mailto:[log in to unmask]] On Behalf Of Benson, Teresa
Sent: Thursday, July 18, 2013 7:47 AM
To: [log in to unmask]
Subject: Using NLP for critical appraisal

I’m wondering if anyone out there has toyed with using natural language recognition/programming software to “read” journal articles and apply elements of a critical appraisal checklist.  While it wouldn’t replace humans, it might be interesting to see if it’s helpful for screening large numbers of articles when doing systematic reviews.  Thanks,

Teresa Benson, MA, LP

Clinical Lead, Evidence-Based Medicine

McKesson Health Solutions
18211 Yorkshire Ave

Prior Lake, MN  55372

[log in to unmask]
Phone: 1-952-226-4033

Confidentiality Notice: This e-mail message, including any attachments, is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message.