Subscribe to:
Post Comments (Atom)
skip to main |
skip to sidebar
About this blog
This is the place to find update notices and discussion around http://thesciencebehindit.net - the site that provides the citations that the journalists couldn't be arsed to include.
Just wondering if it has to be limited to any particular news source. Can it be designed in such a way that any URL (if containing a news piece about science/medicine - otherwise you get "No matches, sorry" response) from anywhere will be possible to use - i.e., your application may be able to extract key terms and make an educated guess as to which paper it was about. Perhaps even blog posts can be input in there, etc.
ReplyDeleteThat would certainly be the ultimate goal. Things are easier with defined news formats because I can pull out dates and so on automatically, and also the styles of writing are consistent enough to help with identifying journals. I think for free text the most effective thing would be to try and match it to a news story from one of the mainstream sources and then work off the features in that - but I admit that's just a hunch and I'm going to have to try it and see.
ReplyDeletenytimes.com
ReplyDeleteI was just looking at The Guardian's website (eg http://www.guardian.co.uk/music/2009/apr/21/study-finds-pirates-buy-more-music) and hopped over here to see if you had set up a sciencebehindit for that.
ReplyDelete