Concordance

Query features 37 >
Shuffle 37 (1,725.0 per million)

other,66-1-N03-1033,bq	fine-grained modeling of <term> unknown word	features	</term> . Using these ideas together , the	#2977 We present a new part-of-speech tagger that demonstrates the following ideas: (i) explicit use of both preceding and following tag contexts via a dependency network representation, (ii) broad use of lexical features, including jointly conditioning on multiple consecutive words, (iii) effective use of priors in conditional loglinear models, and (iv) fine-grained modeling of unknown word features.
other,21-2-E06-1022,bq	utterance </term> and <term> conversational context	features	</term> . Then , we explore whether information	#10276 First, we investigate how well the addressee of a dialogue act can be predicted based on gaze, utterance and conversational context features.
other,6-3-C04-1068,bq	</term> . In this paper , we identify <term>	features	</term> of <term> electronic discussions </term>	#5431 In this paper, we identifyfeatures of electronic discussions that influence the clustering process, and offer a filtering mechanism that removes undesirable influences.
other,6-4-N04-1024,bq	support vector machine </term> uses these <term>	features	</term> to capture <term> breakdowns in coherence	#6711 A support vector machine uses thesefeatures to capture breakdowns in coherence due to relatedness to the essay question and relatedness between discourse elements.
other,27-3-P05-1069,bq	model score </term> ) as well as <term> binary	features	</term> based on the <term> block </term> identities	#9613 We use a maximum likelihood criterion to train a log-linear block bigram model which uses real-valued features (e.g. a language model score) as well as binary features based on the block identities themselves, e.g. block bigram features.
other,8-1-P86-1038,bq	</term> use structures containing sets of <term>	features	</term> to describe <term> linguistic objects	#14631 Unification-based grammar formalisms use structures containing sets offeatures to describe linguistic objects.
other,12-4-I05-5003,bq	techniques </term> are able to produce useful <term>	features	</term> for <term> paraphrase classification	#8409 Our results show that MT evaluation techniques are able to produce usefulfeatures for paraphrase classification and to a lesser extent entailment.
	</term> to fit . One of the distinguishing	features	of a more <term> linguistically sophisticated	#20006 One of the distinguishing features of a more linguistically sophisticated representation of documents over a word set based representation of them is that linguistically sophisticated units are more frequently individually good predictors of document descriptors (keywords) than single words are.
other,8-4-P05-1069,bq	</term> can easily handle millions of <term>	features	</term> . The best system obtains a 18.6	#9634 Our training algorithm can easily handle millions offeatures.
other,51-5-E06-1035,bq	<term> lexical-cohesion and conversational	features	</term> performs best , and ( 3 ) <term> conversational	#10580 Examination of the effect of features shows that predicting top-level and predicting subtopic boundaries are two distinct tasks: (1) for predicting subtopic boundaries, the lexical cohesion-based approach alone can achieve competitive results, (2) for predicting top-level boundaries, the machine learning approach that combines lexical-cohesion and conversational features performs best, and (3) conversational cues, such as cue phrases and overlapping speech, are better indicators for the top-level prediction task.
other,18-1-P06-2012,bq	use of various <term> lexical and syntactic	features	</term> from the <term> contexts </term> . It	#11337 This paper presents an unsupervised learning approach to disambiguate various relations between name entities by use of various lexical and syntactic features from the contexts.
other,23-7-J05-1003,bq	evidence from an additional 500,000 <term>	features	</term> over <term> parse trees </term> that	#8822 The method combined the log-likelihood under a baseline model (that of Collins [1999]) with evidence from an additional 500,000features over parse trees that were not included in the original model.
other,19-4-J05-1003,bq	represented as an arbitrary set of <term>	features	</term> , without concerns about how these	#8729 The strength of our approach is that it allows a tree to be represented as an arbitrary set offeatures, without concerns about how these features interact or overlap and without the need to define a derivation or a generative model which takes these features into account.
other,7-2-P01-1070,bq	which are built from <term> shallow linguistic	features	</term> of <term> questions </term> , are employed	#2152 These models, which are built from shallow linguistic features of questions, are employed to predict target variables which represent a user's informational goals.
	. In this presentation , we describe the	features	of and <term> requirements </term> for a genuinely	#260 In this presentation, we describe the features of and requirements for a genuinely useful software infrastructure for this purpose.
other,15-3-P05-1069,bq	model </term> which uses <term> real-valued	features	</term> ( e.g. a <term> language model score	#9601 We use a maximum likelihood criterion to train a log-linear block bigram model which uses real-valued features (e.g. a language model score) as well as binary features based on the block identities themselves, e.g. block bigram features.
	identities themselves , e.g. block bigram	features	. Our <term> training algorithm </term> can	#9624 We use a maximum likelihood criterion to train a log-linear block bigram model which uses real-valued features (e.g. a language model score) as well as binary features based on the block identities themselves, e.g. block bigram features.
other,14-3-J05-1003,bq	<term> ranking </term> , using additional <term>	features	</term> of the <term> tree </term> as evidence	#8703 A second model then attempts to improve upon this initial ranking, using additionalfeatures of the tree as evidence.
other,19-6-E06-1035,bq	<term> lexical-cohesion and conversational	features	</term> , but do not change the general preference	#10630 We also find that the transcription errors inevitable in ASR output have a negative impact on models that combine lexical-cohesion and conversational features, but do not change the general preference of approach for the two tasks.
other,13-3-C04-1035,bq	create a set of <term> domain independent	features	</term> to annotate an input <term> dataset	#5197 We then use the predicates of such clauses to create a set of domain independent features to annotate an input dataset, and run two different machine learning algorithms: SLIPPER, a rule-based learning algorithm, and TiMBL, a memory-based system.


	in Help