Research Article
Efficient and Effective Learning of HMMs Based on Identification of Hidden States
Figure 13
Probability approximation with large probabilities without mixing states.
(a) Approximation by a first-order HMM assumption |
(b) Approximation by a second-order HMM assumption |