Báo cáo khoa học: "Using Conditional Random Fields For Sentence Boundary Detection In Speech" potx

... Proceedings of the 43rd Annual Meeting of the ACL, pages 451–458, Ann Arbor, June 2005. c 2005 Association for Computational Linguistics Using Conditional Random Fields For Sentence Boundary Detection ... not an- notated according to the guideline used for the train- ing and test data (Strassel, 2003). For BN, we use the training corpus for the LM for speech recogni-...

Ngày tải lên: 31/03/2014, 03:20

... CRF:All in Table 5). We get about 0.5% increase in accuracy, 76.1% with a window of size w = 1. Using larger windows resulted in minor increases in the performance of the model, as summarized in Table ... For our textual phonological features, we in- cluded the number of syllables in a word and the number of phones (both in citation form as well as transcribed form). Instead...

Ngày tải lên: 08/03/2014, 04:22

... linguistic expression used by a ques- tioner to request information in the form of an an- swer. The sentence containing request focus is called question. Context are the sentences contain- ing ... on forum data. Experimental results show that 1) Linear CRFs out- perform SVM and decision tree in both context and answer detection; 2) Skip-chain CRFs outper- form Linear CRFs for answer...

Ngày tải lên: 23/03/2014, 17:20

... experiment simply for illustrative purposes: many columns in this code were unnecessary, yielding only a slight gain in performance over much simpler codes while incurring a very large increase in training ... errors by individual models. Error-correcting CRF training is much less resource intensive and has a much faster training time than a standardly formulated CRF, while decoding p...

Ngày tải lên: 31/03/2014, 03:20

... CRF training. 1 Introduction Conditional random fields (CRFs) are a recently introduced formalism (Lafferty et al., 2001) for representing a conditional model p(y|x), where both a set of inputs, ... isozaki}@cslab.kecl.ntt.co.jp Abstract This paper proposes a framework for train- ing Conditional Random Fields (CRFs) to optimize multivariate evaluation mea- sures, including non-...

Ngày tải lên: 17/03/2014, 04:20

... Effect of Training Data Size In order to allow for rapid examination of multi- ple feature combinations, we restricted the size of the training set (S) to maintain manageable train- ing times. ... training data. We test this assumption by taking the best-performing feature sets from Table 5 and training new models using twice the training data {S=4000}. The results are shown in Table 6....

Ngày tải lên: 17/03/2014, 00:20

... distributions in Figure 1, except the user model, are known. The ASR confusion model was esti- mated by transcribing 50 randomly chosen dialogs from the training set in Section 4.2 and calculat- ing the ... reasonable, hand-crafted value for θ, and then generated syn- thetic dialogs by following the probabilistic process depicted in Figure 1. In this way, we were able to create synt...

Ngày tải lên: 20/02/2014, 09:20

... sometimes elim- inate the original meaning by incorrectly removing important parts of sentences, be- cause trimming probabilities only depend on parents’ and daughters’ non-terminals in applied CFG ... tree of the original sentence is marked if it exists in the compressed sentence. In the figure, the marked terminals are represented by circles. Second, each non-terminal in the origi...

Ngày tải lên: 20/02/2014, 12:20

... word in WordNet is linked to a set of senses, with each sense identifying one particular meaning of that word. For example, the noun ‘skin’ has senses rep- resenting (i) the cutis or skin of ... sense-pair, using the first term of the score tuple as the main key for comparison (lines 14 and 15), and using the second term as a tie-breaker (lines 16 to 18). If the PRO score for relation...

Ngày tải lên: 08/03/2014, 02:21

... parses in TreeBank Release 2. As named entity information is not available in PropBank/TreeBank we tagged the training corpus with NE information using an open-domain NE recognizer, having 96% ... were mainly responsible for the improvements. This is the rationale for including them in the FS2 set. We also were interested in comparing the results 2 3 These results, listed also on...

Ngày tải lên: 08/03/2014, 04:22