Hackday 2007 Entry
Posted by mattc at Jun 17, 07 02:10 PM ... Comments (0)
Although I was unable to turn up on the Sunday, I thought I would tidy up a couple of hours worth of Saturday's hacking effort in to something functional.
The idea, entitled A Cloud of Teenage Angst, experiments with a way of nested tag clouds (aka. weighted lists) inside each other to clarify the meaning of the terms.
It uses the data from the weekly BBC Slink agony aunt column.
The main cloud is formed from the keywords that the BBC editorial team have associated with each column. Selecting a term in this cloud will result in a mini-cloud being inserted in to the document comprising of a frequency analysis of words in all agony columns containing the parent term.
Because the words most often used in free text tend to be the canonical ones (rather than slang, abbreviations etc.) we can create a fairly accurate nested cloud that better describes unfamiliar, abstract, or duplicte entries in the main cloud.
Would be interested in trying this on bigger sets of data.
Comments (0)
Post Your Comments
random bookmark
link summary
month
October 2009 (1)
September 2009 (14)
August 2009 (16)
July 2009 (21)
June 2009 (24)
May 2009 (16)
April 2009 (2)
March 2009 (22)
February 2009 (11)
January 2009 (11)
December 2008 (9)
November 2008 (16)
October 2008 (18)
September 2008 (11)
August 2008 (12)
July 2008 (20)
June 2008 (15)
May 2008 (27)
April 2008 (9)
March 2008 (10)
February 2008 (8)
January 2008 (8)
December 2007 (12)
November 2007 (10)
October 2007 (10)
September 2007 (6)
August 2007 (13)
July 2007 (8)
June 2007 (10)
May 2007 (12)
April 2007 (5)
March 2007 (12)
February 2007 (13)
January 2007 (22)
December 2006 (21)
November 2006 (28)
August 2006 (1)
category
code (15)
food (4)
notes (4)
photo (18)
project (2)
quote (12)
sketch (13)
soup (10)
travel (2)
matt chadburn . wiki . 2006