This is a hard problem. If you go by degrees and shades of synonymity, it can (and has been) done manually - see Visual Thesaurus (http:// www.visualthesaurus.com/).
But for grouping based on the same semantic topics - that's pretty difficult. You could do it based on co-location in a corpus, but there'd be a lot of noise and results wouldn't be perfect. Maybe by using a syntactic parser to pull out nouns and verbs you could improve the results a bit. You could probably get something. Your best bet is definitely a statistical, corpus-linguistics based approach.. Actual semantic parsing isn't anywhere close, yet. The consensus of those I've talked to seems to be that we won't have "full" semantic language processing until we have strong AI - in fact, the two are very probably equivalent. Natural language is just too full of ambiguity and context. For example, there's probably a great many *people* in the world who'd have a hard time linking the words "fish" and "reel", if they haven't fished before. And a lot of people would probably associate "reel" with "dance" more strongly than "fish". The task potentially would strain even the verbal prowess of a non- linguist human, and computers are much more limited. On Jul 28, 4:58 pm, Daniel <doubleagen...@gmail.com> wrote: > I want to write a clojure program that searches for similarities of > words in the english language and places them in a graph, where the > distance between nodes indicates their similarity. I don't mean > syntactical similarity. Related contextual meaning is closer to the > mark. > > For instance: "fish" and "reel" don't have much similarity, but in the > context of fishing they do, so the distance in such a graph wouldn't > be very large. > > I'm sure research has been done in this area (I suspect with no small > portion belonging to google), so can anybody point me in the right > direction? > > Thanks. -- You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com Note that posts from new members are moderated - please be patient with your first post. To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en
