On 04/20/2012 03:29 PM, Jim - FooBar(); wrote:
Hmmm, that sounds like it should work....however you don't want to separate your entities to Street, Town, Province, Post Code, Country etc cos then how are you going to join them to get your 'real' entity (address)? I would say keep the whole address as 1 entity and produce some training data that mark the whole thing...of course if you already have some training is better otherwise you will spend a bit of time creating your annotated corpus...
For me it sounds like he has already a string which contains the address. He just wants to get the individual parts of the address recognized, right? I would train one name finder and let him recognize all the types. Have a look at our documentation on how to do that, if the results do not work out of the box, you can easily adapt the feature generation a bit for your needs, e.g using a town dictionary, street dictionary, etc. Jörn
