Brief Paper: Augmentation of Hidden Markov Chain for Complex Sequential Data in Context

The classical HMM is defined by a parameter triple λ = (π, A, B), where each represents collection of probability distributions: initial state, state transition and output distributions in order. This paper proposes new stationary e =(e1,e2,…,eN) N the number states et P(|xt i,y) for describing how an input pattern y ends xt i at time t followed nothing. It often said that all well well. We argue here should end sets framework theory presents efficient inference training algorithms based on dynamic programming expectation-maximization. proposed model applicable to analyzing any sequential data with two or more finite segmental patterns are concatenated, forming context its neighbors. Experiments online Hangul handwriting characters have proven effect augmentation terms highly intuitive segmentation as recognition performance 13.2% error rate reduction.