Concordance

Query training 34 (1,585.2 per million)

	and a new methodology for automatically	training	<term> SPoT </term> on the basis of <term> feedback	#1353 In this paper, we present SPoT, a sentence planner, and a new methodology for automatically training SPoT on the basis of feedback provided by human judges.
lr,8-6-N01-1003,bq	rules </term> automatically learned from <term>	training	data </term> . We show that the trained <term>	#1430 The SPR uses ranking rules automatically learned fromtraining data.
tech,0-1-P01-1056,bq	</term> . <term> Techniques for automatically	training	</term> modules of a <term> natural language	#2015 Techniques for automatically training modules of a natural language generator have recently been proposed, but a fundamental concern is whether the quality of utterances produced with trainable components can compete with hand-crafted template-based or rule-based approaches.
other,18-3-P01-1070,bq	including the influence of various <term>	training	and testing factors </term> on <term> predictive	#2188 We report on different aspects of the predictive performance of our models, including the influence of varioustraining and testing factors on predictive performance, and examine the relationships among the target variables.
lr,15-1-N03-1001,bq	<term> manual transcription </term> of <term>	training	data </term> . The method combines <term> domain	#2220 This paper describes a method for utterance classification that does not require manual transcription oftraining data.
tech,4-3-N03-1001,bq	</term> . In our method , <term> unsupervised	training	</term> is first used to train a <term> phone	#2260 In our method, unsupervised training is first used to train a phone n-gram model for a particular domain; the output of recognition with this model is then passed to a phone-string classifier.
lr,2-1-N03-2003,bq	<term> tagging </term> result . Sources of <term>	training	data </term> suitable for <term> language modeling	#3016 Sources oftraining data suitable for language modeling of conversational speech are limited.
lr,7-2-N03-2003,bq	limited . In this paper , we show how <term>	training	data </term> can be supplemented with <term>	#3035 In this paper, we show howtraining data can be supplemented with text from the web filtered to match the style and/or topic of the target recognition task, but also that it is possible to get bigger performance gains from the data by using class-dependent interpolation of N-grams.
	bootstrapping procedure </term> is implemented as	training	two <term> successive learners </term> . First	#3340 The bootstrapping procedure is implemented as training two successive learners.
tech,1-4-N03-2036,bq	trigram language model </term> . During <term>	training	</term> , the <term> blocks </term> are learned	#3445 Duringtraining, the blocks are learned from source interval projections using an underlying word alignment.
lr,27-2-P03-1050,bq	<term> parallel corpus </term> as its sole <term>	training	resources </term> . No <term> parallel text	#4473 The stemming model is based on statistical machine translation and it uses an English stemmer and a small (10K sentences) parallel corpus as its soletraining resources.
other,7-3-P03-1050,bq	parallel text </term> is needed after the <term>	training	phase </term> . <term> Monolingual , unannotated	#4483 No parallel text is needed after thetraining phase.
lr,34-5-P03-1051,bq	expanded <term> vocabulary </term> and <term>	training	corpus </term> . The resulting <term> Arabic	#4740 To improve the segmentation accuracy, we use an unsupervised algorithm for automatically acquiring new stems from a 155 million word unsegmented corpus, and re-estimate the model parameters with the expanded vocabulary andtraining corpus.
lr,11-2-P03-1058,bq	automatically acquire <term> sense-tagged	training	data </term> from <term> English-Chinese parallel	#4833 In this paper, we evaluate an approach to automatically acquire sense-tagged training data from English-Chinese parallel corpora, which are then used for disambiguating the nouns in the SENSEVAL-2 English lexical sample task.
tech,24-3-C04-1080,bq	</term> , we present a method of <term> HMM	training	</term> that improves <term> accuracy </term>	#5583 Observing that the quality of the lexicon greatly impacts the accuracy that can be achieved by the algorithms, we present a method of HMM training that improves accuracy when training of lexical probabilities is unstable.
	</term> that improves <term> accuracy </term> when	training	of <term> lexical probabilities </term> is	#5588 Observing that the quality of the lexicon greatly impacts the accuracy that can be achieved by the algorithms, we present a method of HMM training that improves accuracy when training of lexical probabilities is unstable.
lr,24-3-C04-1112,bq	classifier </term> , therefore augmenting the <term>	training	material </term> available to the algorithm	#6048 The advantage of this novel method is that it clusters all inflected forms of an ambiguous word in one classifier, therefore augmenting thetraining material available to the algorithm.
lr,13-1-H05-1012,bq	Arabic-English </term> based on <term> supervised	training	data </term> . We demonstrate that it is	#7263 This paper presents a maximum entropy word alignment algorithm for Arabic-English based on supervised training data.
lr,8-2-H05-1012,bq	demonstrate that it is feasible to create <term>	training	material </term> for problems in <term> machine	#7274 We demonstrate that it is feasible to createtraining material for problems in machine translation and that a mixture of supervised and unsupervised methods yields superior performance.
tech,16-3-H05-1095,bq	<term> phrases </term> , as well as a <term>	training	method </term> based on the maximization	#7386 A statistical translation model is also presented that deals such phrases, as well as atraining method based on the maximization of translation accuracy, as measured with the NIST evaluation metric.


	in Help