One of the take-aways of all of this work is that it is difficult to get large amounts of performance lift from resources built for purposes different than what you are doing.
Wordnet was not built to specifically help with search or with data mining. It was built because people wanted to explore whether a machine readable dictionary like data source would help. Occasionally it does, but more often it does not. The observation that links are not weighted is one problem that is commonly encountered, but that only raises the question of how to weight links. That question has no single answer. On Thu, Jun 18, 2009 at 8:54 AM, Stephen Green <[email protected]>wrote: > Seriously, though, there's been a lot of work on this, and you should get > some good ideas from some of the work that was done in this century :-) > Man, I'm getting old. -- Ted Dunning, CTO DeepDyve
