Báo cáo khoa học: "An Unsupervised Dynamic Bayesian Network Approach to Measuring Speech Style Accommodation" doc

... describe the speech features used, and the proposed mod- els. 3.1 Dynamic Bayesian Networks The theory of Bayesian networks is well doc- umented and understood (Jensen, 1996; Pearl, 1988). A Bayesian network ... 27 2012. c 2012 Association for Computational Linguistics An Unsupervised Dynamic Bayesian Network Approach to Measuring Speech Style Accommodation Ma...

Ngày tải lên: 31/03/2014, 21:20

... of Distributed Listening research is to take a unique approach in order to enhance the success of the traditional approaches to speech recognition. The approach of Distributed Listen- ing directly ... with multiple speech recognizers in an effort to improve speech recognition, as discussed next. 2.1 Enhanced Majority Rules Barry (et al., 1994) took three different Autom...

Ngày tải lên: 08/03/2014, 01:20

... HIER first attempts to generate the full sentence as a single phrase from P t , then falls back to ITG -style derivations to cope with sparsity. We allow for this within the Bayesian ITG context by ... distribution with the average length parameter λ. As long phrases lead to spar- sity, we set λ to a relatively small value to allow us to bias against overly long phrases 4 . P...

Ngày tải lên: 20/02/2014, 04:20

... word pairs that are not likely to be good predictors of discourse relations. To test this hypothesis, we decided to carry out a second experiment that used as predictors only a subset of the word ... us access to general purpose knowledge bases that assert that “similar(libya,rwanda)”. The approach we advocate in this paper is in some respects less ambitious than current approaches...

Ngày tải lên: 20/02/2014, 21:20

... unsupervised corpus-based approach to prepositional phrase attachment that outperforms previous unsupervised techniques and approaches the performance of supervised methods. Unlike previous unsupervised ... University of Manitoba. Abstract Prepositional phrase attachment is a common source of ambiguity in natural language processing. We present an unsupervised corpus-based approa...

Ngày tải lên: 08/03/2014, 05:20

... is similar to words extraction in automatic speech recogni- tion (ASR)(Jurafsky and Martin, 2000, chp. 5,7). The states of the ASR model are phones, where each observation is a vector of spectral ... supervised training of a speech recognizer combines a training corpus of speech wave files, together with word-transcription, and language model probabilities, in order to learn the phon...

Ngày tải lên: 23/03/2014, 18:20

... target concepts in a set of documents. The sys- tem then learns a model of the extraction task, which can be applied to new documents auto- matically. Both of these approaches require massive ... or more arbitrary tokens, and slots match instance attributes. The limited skips match zero or more arbitrary tokens, which must not belong to entities of the types equal to the types of...

Ngày tải lên: 23/03/2014, 18:20

... data. Domain Tokens Types TTR Internet Total 15919 4152 0.26 English 963 283 0.29 Space Total 16066 3938 0.25 English 485 73 0.15 EU Total 16028 4048 0.25 English 49 30 0.61 Table 1: English token and ... required to recognise them. Our aim is to address these issues in future work. 4 System Description Our system is a UNIX pipeline which converts HTML documents to XML and applies a set...

Ngày tải lên: 23/03/2014, 19:20

... salad of lettuce and tomatoes 'Lettuce' attaches low to 'tomatoes', giving: L-attach: (salad (of ((lettuce) and (tomatoes))) [AR98] models. In addition to these, a corpus- ... helpful to compare its performance to that of an analogous supervised model. The smaller the error reduction when going from unsupervised to supervised models, the more comparabl...

Ngày tải lên: 23/03/2014, 19:20

... con- cept vector is the closest to the instance vector to the target word. To create the the instance vector, we use the words that occur in the same abstract as the target word as features. To create ... paper, we introduce an unsupervised vector approach to disambiguate words in biomedical text using contextual information from the UMLS and Medline. We compare our approach...

Ngày tải lên: 31/03/2014, 00:20