Wikipedia for teaching and research
Posted by Tim Stevens on 12 May 2008
Hot off the back of news that ‘Once shunned by academics, Wikipedia now a teaching tool‘ comes the launch of semantic search enginge Powerset, whose ‘natural language search technology … has read 2.5 million Wikipedia pages and extracted “meaning” from the sentences, creating a navigation and semantic layer on top of the popular Web encyclopedia.’
I’ve been playing around with the beta this morning and I like it. The results are far more ‘intelligent’ than the rather clunky Wikipedia search engine which generally returns zero/lame results unless you get the search criteria exactly right. Powerset has also indexed Freebase, Metaweb’s open structured database. Makes me wonder what happened to Google’s prototype search engine project Searchmash, which always returned a far wider assortment of relevant links whenever I used it.
As a bit of fun, you can also monitor your own web footprint at Addict-o-matic, a quick and easy way of seeing where your site shows up across various web aggregators.
12 May 2008 at 2:01 pm
Sir, thank you for this interesting post.
Have you seen this article in The New Yorker, “Virtual Iraq”?
http://www.newyorker.com/reporting/2008/05/19/080519fa_fact_halpern
12 May 2008 at 5:30 pm
Hi Ortho,
Thanks for the link. I’ve read various articles on ‘Virtual Iraq’ but not seen that one. In typical New Yorker fashion it’s a lot smarter than most.