I'm trying to understand if Markov models can account for a "noise event" when predicting the next item in a sequence.

For instance, if i have very frequently occurring (noise) event "F", can a markov model ignore the presence of such "noise events". So the following two sequences must be considered equivalent:

B -> A -> C

B -> A -> F -> C

I know I could just filter out the "F" event, but I don't know these in advance. Moreover, the sequence B -> A -> F could also be an "interesting" sequence, and I might want to predict that "F" occurred after B & A did.


Your Answer

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

Browse other questions tagged or ask your own question.