Access the full text.
Sign up today, get DeepDyve free for 14 days.
I. Myung (2003)
Tutorial on maximum likelihood estimationJournal of Mathematical Psychology, 47
B. Sin, J. Kim (1995)
Nonstationary hidden Markov modelSignal Process., 46
Yujian Li (2005)
Hidden Markov models with states depending on observationsPattern Recognit. Lett., 26
Xiao Xi (2004)
Duration Distribution Based HMM Speech Recognition Models
R. Rosenfeld (1996)
A maximum entropy approach to adaptive statistical language modellingComput. Speech Lang., 10
Nonstationary Maximum Entropy Markov Model and the Pos-Tagging Task
Guanglu Sun, C. Huang, Xiaolong Wang, Zhi-ming Xu (2006)
Chinese Chunking Based on Maximum Entropy Markov ModelsInt. J. Comput. Linguistics Chin. Lang. Process., 11
J. Lafferty, A. McCallum, Fernando Pereira (2001)
Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data
Sven Martin, H. Ney, Jörg Zaplo (1999)
Smoothing methods in maximum entropy language modeling1999 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings. ICASSP99 (Cat. No.99CH36258), 1
(2006)
Received July
S. Pietra, V. Pietra, J. Lafferty (1995)
Inducing Features of Random FieldsArXiv, abs/cmp-lg/9506014
R. Rosenfeld (2001)
A Maximum Entropy Approach to Adaptive Statistical Language Modeling
T. Cover, Joy Thomas (1991)
Elements of Information Theory
P. Brown, V. Pietra, P. Souza, J. Lai, R. Mercer (1992)
Class-Based n-gram Models of Natural LanguageComput. Linguistics, 18
M. Collins, Brian Roark, M. Saraçlar (2005)
Discriminative Syntactic Language Modeling for Speech Recognition
H. Ng, Jin Low (2004)
Chinese Part-of-Speech Tagging: One-at-a-Time or All-at-Once? Word-Based or Character-Based?
J. Darroch, D. Ratcliff (1972)
Generalized Iterative Scaling for Log-Linear ModelsAnnals of Mathematical Statistics, 43
Issam Bazzi, R. Schwartz, J. Makhoul (1999)
An Omnifont Open-Vocabulary OCR System for English and ArabicIEEE Trans. Pattern Anal. Mach. Intell., 21
Q. Dong, Xiaolong Wang, Lei Lin, Yi Guan (2005)
A seqlet-based maximum entropy Markov approach for protein secondary structure predictionScience in China Series C: Life Sciences, 48
I. Good (1953)
THE POPULATION FREQUENCIES OF SPECIES AND THE ESTIMATION OF POPULATION PARAMETERSBiometrika, 40
(2007)
ACM Trans. Asian Language Information Processing Pub. Date
Jinghui Xiao, Bingquan Liu, Xiaolong Wang (2005)
Principles of Non-stationary Hidden Markov Model and Its Applications to Sequence Labeling Task
Jianfeng Gao, Hao Yu, Wei Yuan, P. Xu (2005)
Minimum Sample Risk Methods for Language Modeling
Stanley Chen, R. Rosenfeld (2000)
A survey of smoothing techniques for ME modelsIEEE Trans. Speech Audio Process., 8
(1980)
Variable duration models for speech
L. Rabiner (1989)
A tutorial on hidden Markov models and selected applications in speech recognitionProc. IEEE, 77
A. McCallum, Dayne Freitag, Fernando Pereira (2000)
Maximum Entropy Markov Models for Information Extraction and Segmentation
(1999)
Foundation of Statistic Natural Language Processing
H. Sameti, L. Deng (2002)
Nonstationary-state hidden Markov model representation of speech signals for speech enhancementSignal Process., 82
P. Djurić, Joon-Hwa Chun (2002)
An MCMC sampling approach to estimation of nonstationary hidden Markov modelsIEEE Trans. Signal Process., 50
Sequence labeling is a core task in natural language processing. The maximum entropy Markov model (MEMM) is a powerful tool in performing this task. This article enhances the traditional MEMM by exploiting the positional information of language elements. The stationary hypothesis is relaxed in MEMM, and the nonstationary MEMM (NS-MEMM) is proposed. Several related issues are discussed in detail, including the representation of positional information, NS-MEMM implementation, smoothing techniques, and the space complexity issue. Furthermore, the asymmetric NS-MEMM presents a more flexible way to exploit positional information. In the experiments, NS-MEMM is evaluated on both the Chinese and the English pos-tagging tasks. According to the experimental results, NS-MEMM yields effective improvements over MEMM by exploiting positional information. The smoothing techniques in this article effectively solve the NS-MEMM data-sparseness problem; the asymmetric NS-MEMM is also an improvement by exploiting positional information in a more flexible way.
ACM Transactions on Asian Language Information Processing (TALIP) – Association for Computing Machinery
Published: Sep 1, 2007
Read and print from thousands of top scholarly journals.
Already have an account? Log in
Bookmark this article. You can see your Bookmarks on your DeepDyve Library.
To save an article, log in first, or sign up for a DeepDyve account if you don’t already have one.
Copy and paste the desired citation format or use the link below to download a file formatted for EndNote
Access the full text.
Sign up today, get DeepDyve free for 14 days.
All DeepDyve websites use cookies to improve your online experience. They were placed on your computer when you launched this website. You can change your cookie settings through your browser.