HMM在生物信息中的应用2.pptVIP

HMMs及其应用Hidden Markov Models Its Application Lecture # 11 Four Basic Problems for HMMs Determine the structure/topology of the HMM—states and transitions Determine the values of the model parameters—emission and transition probabilities (the learning problem). That is, how to adjust the model parameters to maximize P(O | m)? Given an observation sequence O = O1O2O3…Ot, and a model m = (T, E, ?), how to evaluate the probability of this sequence given the model (the estimation problem). That is, how to efficiently compute P(O | m)? Given an observation sequence O and a model m, how to find the