... parts randomly: 5K as the adaptation corpusand 5K as the testing set. We show the ASR char-acter accuracy results after lexicon adaptation bythe proposed approach in Table 3.LAICA-1 LAICA-2 A ... replaced by characters, we cantreat words as a means to enhance character recog-nition accuracy. Such arguments stand at least forChinese ASR since they evaluate on character errorrate and ... total path probability mass. This can beamended by involving the discriminative language model adaptation in the iteration, which results in a unified languagemodel and lexicon adaptationframework....
... Lexicon Models for Statistical Machine Translation using a MaximumEntropy ApproachIsmael Garc´ a VareaDpto. de Inform´aticaUniv. de Castilla-La ManchaCampus Universitario s/n02071 Albacete, ... dialog act.To include this additional information within thestatistical framework we use the maximum en-tropy approach. This approach has been appliedin natural language processing to a variety ... candide system for machine translation.In Proc. , ARPA Workshop on Human Language Technology, pages 157–162.Adam L. Berger, Stephen A. Della Pietra, and Vin-cent J. Della Pietra. 1996. A maximum...
... Abdullah, M. B., Al-Nasser, A. D. & Nooreha, H. 2000. Evaluating Functional Relationship Between Image, Customer Satisfaction and Loyalty using General maximum Entropy. Total Quality Management ... diagram to analyse a set of relationships between variables. It differs from simple path analysis in that all variables are latent variables measured by multiple indicators, which have associated ... relations among exogenous variables ( i.e; a variables that is not caused by another variable in the model) , and endogenous variables (i.e; a variables that is caused by one or more variable...
... signif-icantly. Bear in mind that Charniak et al. (2003) in-tegrated Charniak’s languagemodel with the syntax-based translation model Yamada and Knight pro-posed (2001) to rescore a tree-to-string ... Stochastic analysis of lexical andsemantic enhanced structural language model. The 8thInternational Colloquium on Grammatical Inference(ICGI), 97-111.K. Yamada and K. Knight. 2001. A syntax-based ... (EMNLP),858-867.E. Charniak. 2001. Immediate-head parsing for language models. The 39th Annual Conference on Associationof Computational Linguistics (ACL), 124-131.E. Charniak, K. Knight and K. Yamada. 2003....
... and Linda C. Bauman Peto. 1995. A hierarchical Dirichlet language model. Natural Lan-guage Engineering, 1(3):1–19.Y.W. Teh. 2006. A hierarchical Bayesian language model based on Pitman-Yor processes. ... n-grams:C(ab) − C(ab∗). A( ab) = max(1, K(C(ab) − C(ab∗))) A different K constant is chosen for each n-gramorder. Using this formulation as an interpolated 5-gram languagemodel gives a cross entropy ... Speech and Language. R. Kneser and H. Ney. 1995. Improved backing-off form-gram language modeling. In International Confer-ence on Acoustics, Speech, and Signal Processing.David J. C. Mackay and...
... com-pression tasks achieved a significant com-pression rate without any loss.1 IntroductionThere has been an increase in available N -gramdata and a large amount of web-scaled N-gramdata has been ... the ACL-IJCNLP 2009 Conference Short Papers, pages 341–344,Suntec, Singapore, 4 August 2009.c2009 ACL and AFNLP A Succinct N-gram Language Model Taro Watanabe Hajime Tsukada Hideki IsozakiNTT ... Communication Science Laboratories2-4 Hikaridai Seika-cho Soraku-gun Kyoto 619-0237 Japan{taro,tsukada,isozaki}@cslab.kecl.ntt.co.jpAbstractEfficient processing of tera-scale text datais an important...
... NIST Language Recognition Evaluation database. 1 Introduction Spoken language and written language are similar in many ways. Therefore, much of the research in spoken language identification, ... Recognition Evaluation (LRE) data. The database was intended to establish a baseline of performance capability for language recognition of conversational tele-phone speech. The database contains recorded ... by a chan-nel noise. The n-gram languagemodel has achieved equal amounts of success in both tasks, e.g. n-character slice for text categorization by lan-guage (Cavnar and Trenkle, 1994) and...
... Proceedings of the Human Language Technology Workshop, 272-277. ARPA. Raymond Lau, Ronald Rosenfeld, and Salim Roukos. 1993. Trigger-based language models: a maximum entropy approach. In Proceedings ... University, Baltimore, MD. Frederick Jelinek, John Lafferty, David M. Mager- man, Robert Mercer, Adwait Ratnaparkhi, Salim Roukos. 1994. Decision Tree Parsing using a Hid- den Derivational Model. ... those assigned man- ually in the Penn Treebank (Marcus95) after under- going headword percolation and binarization. All four LMs predict a word wk and they were implemented using the Maximum...
... Ducharme, P. Vincent, and C. Jauvin. 2003. A Neural Probabilistic Language Model. Journal ofMachine Learning Research, 3:1137–1155. A. Berger, V. Della Pietra, and S. Della Pietra. 1996. A Maximum ... Language, 8:1–38.B. Roark, M. Saraclar, and M. Collins. 2007. Discrimi-native n-gram Language Modeling. Computer, Speechand Language, 21:373–392.R. Rosenfeld. 1994. Adaptive Statistical Language ... Statistical Language Mod-elling: AMaximumEntropy Approach. Ph.D. thesis,Carnegie Mellon University.R. Rosenfeld. 1996. AMaximumEntropy Approach toAdaptive Statistical Language Modeling. Computer,Speech...
... the small data track. Both thebilingual training data and the trigram language model training data are restricted to the suppliedcorpus, which contains 20k sentences, 179k Chi-nese words and ... 100080{liuqun, sxlin}@ict.ac.cnAbstractWe propose a novel reordering model forphrase-based statistical machine transla-tion (SMT) that uses amaximum entropy (MaxEnt) model to predicate reorderingsof ... (phrase pairs). The model provides content-dependent, hier-archical phrasal reordering with general-ization based on features automaticallylearned from a real-world bitext. Wepresent an algorithm...
... indepen-dently in training.3 Discriminative LanguageModel withPseudo-Negative samplesWe propose a novel discriminative language model; a Discriminative LanguageModel with Pseudo-Negative samples (DLM-PN). ... spe-cific applications and therefore were able to obtainreal negative examples easily. For example, Roark(2007) proposed a discriminative language model, inwhich amodel is trained so that a correct ... June.Brian Roark, Murat Saraclar, and Michael Collins. 2007.Discriminative n-gram language modeling. computerspeech and language. Computer Speech and Lan-guage, 21(2):373–392.Roni Rosenfeld, Stanley...
... Lee, L. S. et al. (1990). A Mandarin Dictation Machine Based Upon A Hierarchical Recognition Approach and Chinese Natural Language Analysis, IEEE Trans. on Pattern Analysis and Machine Intelligence, ... unification granunar and Markov languagemodel are integrated in a word lattice parsing algorithm based on an augmented chart, and the island-driven parsing concept is combined with various ... correct rate of recognition can be as high as 98.3%. This indicates that the language processor based on the integration of the unification grammar and the Markov languagemodel can in fact be...
... features, as described below: a statistical languagemodel and a measure of tensedifficulty.4.1 The language model The lexical difficulty of a text is quite an elaboratephenomenon to parameterise. ... poems as outliers).4 Selection of lexical and syntacticvariablesAny text classification tasks require an object(here a text) to be parameterised into variables,whether qualitative or quantitative. ... Belgiumthomas.francois@uclouvain.beAbstractReading is known to be an essential taskin language learning, but finding the ap-propriate text for every learner is far fromeasy. In this context, automatic...
... Toku- naga, and Hozumi Tanaka. 1998b. A frame- work of integrating syntactic and lexical statis- tics in statistical parsing. Journal of Nat- ural Language Processing, 5(3):85-106. Japanese). ... training data and the accu- racy, we found that good accuracy can be achieved even with a very small set of training data. We believe that the maximumentropy framework has suitable characteristics ... Natural Language Processing, pages 97-106. Adam L. Berger, Stephen A. Della Pietra, and Vincent J. Della Pietra. 1996. Amaximum en- tropy approach to natural language processing. Computational...
... Information Retrieval and Filtering: An Empirical Basis for Grammatical Rules. Information Processing & Management, May. M. Magerman. 1996 Learning Grammatical Struc- ture Using Statistical ... Programs for Machine Learning. San Mateo, CA. Morgan Kaufmann. 3. Richards, D. Landgrebe and P. Swain. 1981 On the accuracy of pixel relaxation labelling. IEEE Transactions on System, Man and Cybernetics. ... All this makes that the performance cannot reach 100%, and that an accurate analysis of the noise in WS3 corpus should be performed to estimate the actual upper bound that a tagger can achieve...