I need to parse international addresses to its individual components (street, city, etc.). After some reasonable survey, I found that HMMs/CRFs are the way to go.
Has anybody had any success using open source implementation of HMMs or CRFs for the address tokenization problem. If yes, what are they? Also, do any implementations provide HMM parameters that they have got after training it for the above problem.
Many thanks in advance.
ps: I have looked at some of the questions like this on SO but none of them request for sophisticated techniques like HMMs, etc. for the problem at hand.