Suggestion tools

In Data on March 23, 2010 by africker




Genius cartoon for Prospect Magazine by

I loved this cartoon from a recent issue of Prospect Magazine (the work of 


I find suggestion tools fascinating – an area where I really wish I had the mathematical skills to get involved. 

The Netflix prize is an interesting example of competitive crowd sourcing to try and improve the recommendation algorithm for a film website. 

Films and book purchases have the advantage of being able to build up a pattern of choices.  This is not without problems (as the comic illustrates) .   Amazon can get very confused by gift purchases (please note for the record Amazon – I am not interested in ER just because I once bought a series on DVD) and has a tendency to offer the same old (I bought CD X might I also be interested in Coldplay? Where X equals pretty much anything).

There was an interesting column on this topic in a recent Word magazine.  It considered how these sorts of tools just tend to recommend more of the same – you like Pavement so you will like the Fall etc etc and outlined a few that throw in the kind of random sidewise steps that can be interesting. 

Closer to professional home there are obvious applications in Pubmed where I notice they now offer both “Also try” with additional search terms and  “Titles with your search terms” where it used to be “Related items”.  The NHS Database Search Tool does not offer anything like this even though the Ovid interface that is the main one hidden underneath does.  NHS Evidence restricts itself to spelling corrections at present.

I wonder if anyone is working on a tool that would offer the sidewise serendipitous type link for health information?


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: