Algorithm of parameter estimation of HMM via Gibbs sampling

Computer Engineering and Applications ›› 2012, Vol. 48 ›› Issue (18): 57-60.

Previous Articles Next Articles

Algorithm of parameter estimation of HMM via Gibbs sampling

ZHU Chengwen1, LI Bing2, HU Kui3

1.Wuhan Ordnance Non-commissioned Officer Academy PLA, Wuhan 430075, China
2.College of Science, National University of Defense Technology, Changsha 410073, China
3.Unit 92941 of PLA, China

Online:2012-06-21 Published:2012-06-20

HMM参数估计的Gibbs抽样算法

朱成文1，李兵2，胡奎3

1.武汉军械士官学校，武汉 430075
2.国防科技大学理学院，长沙 410073
3.中国人民解放军92941部队96分队

Abstract

Abstract: The parameter estimation of Hidden Markov Model（HMM） is critical to all its applications. The classic Baum-Welch algorithm is not flexible with the initial parameters and is easy to fall into the local optimal solution. The great computational power of MCMC is employed. The algorithm of parameter estimation of HMM via Gibbs sampling avoids the local optimal solution and can be more effective.

Key words: Hidden Markov Models（HMM）, Gibbs sampling, conjugate priors

摘要： 隐马氏模型（HMM）的参数估计是隐马氏模型各种应用的关键。经典的Baum-Welch算法容易陷入局部最优，对初始参数的要求苛刻。HMM参数估计的Gibbs抽样法，充分利用模型先验信息，借助马氏链蒙特卡洛方法（MCMC）的强大计算功能，避免了陷入局部最优，有更好的效果。

关键词: 隐马氏模型, Gibbs抽样, 共轭先验

ZHU Chengwen1, LI Bing2, HU Kui3. Algorithm of parameter estimation of HMM via Gibbs sampling[J]. Computer Engineering and Applications, 2012, 48(18): 57-60.

朱成文1，李兵2，胡奎3. HMM参数估计的Gibbs抽样算法[J]. 计算机工程与应用, 2012, 48(18): 57-60.

[1]	SHI Qingwei, WANG Jun, GUO Pengfei. Citation-author-topic evolution model applied in expert retrieval [J]. Computer Engineering and Applications, 2016, 52(6): 55-60.
[2]	WANG Yonggui, ZHANG Xu, LIU Xianguo. Research on micro-blog user’s interest mining based on author-topic model [J]. Computer Engineering and Applications, 2015, 51(13): 126-130.
[3]	XU Hui1, Riyiman TURSUN1，2, Wushour SILAMU2. Online-handwriting recognition research of Uyghur word using GMM and HMM [J]. Computer Engineering and Applications, 2014, 50(11): 202-205.
[4]	FU Ling, ZHANG Hui. Multi-document summary using LDA and spectral clustering [J]. Computer Engineering and Applications, 2013, 49(16): 142-145.
[5]	ELI Jume1，Halidan.A1，HUANG Hao2. Recognition of extracting Uyghur texts from videos images [J]. Computer Engineering and Applications, 2011, 47(36): 190-192.
[6]	JIAO Binliang，CHEN Shuang. Face recognition based on PCA [J]. Computer Engineering and Applications, 2011, 47(18): 201-203.
[7]	YAO Quanzhu，SONG Zhili，PENG Cheng. Research on text categorization based on LDA [J]. Computer Engineering and Applications, 2011, 47(13): 150-153.
[8]	ZHANG Bo-ping¹，WU Song-li²，YAN Jing-feng¹. HMM segmentation method based on statistical layered model for image of vehicle [J]. Computer Engineering and Applications, 2010, 46(5): 162-165.
[9]	HOU Cheng-feng，Gulila·Altenbek. Improved hidden Markov models used in Kazakh part-of-speech tagging [J]. Computer Engineering and Applications, 2010, 46(36): 147-149.
[10]	XIAO Yong¹，QIN Ai-na². Noise robust speech recognition based on improved hidden Markov model and wavelet neural network [J]. Computer Engineering and Applications, 2010, 46(22): 162-164.
[11]	TAO Long-ming,SHI Zhi-cai,PENG Dan,MA Wu. Application of HMM to detecting sophisticated network attacks [J]. Computer Engineering and Applications, 2008, 44(7): 136-138.
[12]	ZHU Wen-qiu,LIU Qiang. Conditional Random Fields with loop and its inference algorithm [J]. Computer Engineering and Applications, 2008, 44(28): 180-183.
[13]	ZHAO Jing¹,ZHANG Qiang¹,WEI Xiao-peng^1,2,ZHOU Shi-hua¹. FLDA，CPCA and HMM for face recognition [J]. Computer Engineering and Applications, 2008, 44(16): 191-193.
[14]	WANG Shuang-cheng^1,2,LENG Cui-ping¹,HOU Cai-hong¹. Naive Bayes method in operational risk level prediction [J]. Computer Engineering and Applications, 2008, 44(12): 26-28.
[15]	LUO Shuang-hu,OUYANG Wei-min. HMM based text categorization [J]. Computer Engineering and Applications, 2007, 43(30): 179-181.

Algorithm of parameter estimation of HMM via Gibbs sampling

HMM参数估计的Gibbs抽样算法

PDF

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 15

Recommended Articles

Metrics