# Hidden markov model to estimate confidence in binary time series

I have binary time series representing active/inactive states

eg. [,1,0,0,1,1,1,1,1,1,1,1,1,1,0,1,1,1,1,1,1,1,1,1,1,1,1,1,0,1,1,0,1,1,1,1,1,0,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,0,1,1,1,1,1,1,1,0,1,0,1,1,1,1,0,1,0,1,1,0,1,1,1,1,1,1,1,1,1,1,1,0,1,1,1,0,1,1,1,1,1,1,1,1,1,1,1,]

These observations of active/inactive are not necessarily real, but do correspond to hidden states of actually being on/off.

Can hidden markov models be used as a binary classifier, in order to compute the likelihood of each new sample in the series, and determine whether the latest observation is likely to be real?