CAJ | 학술논문

隐马尔可夫模型（Hidden Markov Model，HMM）的基本算法体系主要包括Baum-Welch算法、前向-后向算法与Viterbi算法三大经典算法，通过展开对HMM新问题及新算法的理论研究，引出逆维特比问题及逆维特比算法的理论体系，并提出将逆维特比算法引入HMM基本算法体系中构建一种新的算法体系及一种新评估体系的构想，最后对新算法体系的应用进行了展望。
은마이가부모형（Hidden Markov Model，HMM）적기본산법체계주요포괄Baum-Welch산법、전향-후향산법여Viterbi산법삼대경전산법，통과전개대HMM신문제급신산법적이론연구，인출역유특비문제급역유특비산법적이론체계，병제출장역유특비산법인입HMM기본산법체계중구건일충신적산법체계급일충신평고체계적구상，최후대신산법체계적응용진행료전망。
The system of the basic Hidden Markov Model (HMM) algorithm includes the three important algorithms of Baum-Welch algorithm,Forward-Backward algorithm and Viterbi algorithm. The Inv-viterbi problem and its corresponding algorithm is put forward in the paper after the research of the new HMM problem and algorithm. Then a new algorithm system and a new evaluation system are built by introducing the Inv-viterbi algorithm into the HMM basic algorithm system. Finally we give a prospect on the application of the new algorithm system.