Maximum entropy classifier is mostly used for Natural Language Processing, in which the datas are discrete. I learned the formalism from the paper A maximum entropy approach to natural language processing and understood most of it.

My question is, how to apply the Maximum Entropy principle to do the classification on a continuous data set?

I am asking for some sources where I can learn the formalism, not the software packages.

  • $\begingroup$ Pardon my ignorance, but what is a continuous dataset? $\endgroup$ – gregory_britten Jun 24 '14 at 11:54
  • $\begingroup$ @gregory_britten , the value of the features are real floating number. Maybe there are some better phrases, sorry about the confusion. $\endgroup$ – EricCoding Jun 25 '14 at 14:46

Finally I found this paper,

[1] D. Yu, L. Deng, and A. Acero, Pattern Recognit. Lett. 30, 1295–1300 (2009).

That is exactly what I need.

| cite | improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.