Báo cáo khoa học: "An Unsupervised Model for Statistically Determining Coordinate Phrase Attachment" pptx

... statistical model for determining the at- tachment of ambiguous coordinate phrases (CP) of the form nl p n2 cc n3. The model pre- sented here is based on JAR98], an unsupervised model for determining ... An Unsupervised Model for Statistically Determining Coordinate Phrase Attachment Miriam Goldberg Central High School & Dept. of Computer and Informat...

Ngày tải lên: 23/03/2014, 19:20

... reliability of the phrase pair. It will be high for common phrase pairs that are gen- erated directly from the model, and also for phrases that, while not directly included in the model, are composed ... (θ|E, F). (1) If θ takes the form of a scored phrase table, we can use traditional methods for phrase- based SMT to find P (e|f , θ) and concentrate on creating a model for...

Ngày tải lên: 20/02/2014, 04:20

... We were therefore interested in determining the perfor- mance of a trained classifier for our task. We ex- perimented with a conditional Markov model tagger that performed well on language-independent ... POS-tagger 136 does not perform with perfect accuracy particularly on data containing foreign inclusions. Providing the tagger with this information is therefore not neces- sarily usefu...

Ngày tải lên: 23/03/2014, 19:20

... cross-product list. For example, the underscore is defined as a word boundary for English WIL phrases. This assumption is not followed for cer- tain phrases like ”New York” and ”New Mexico”. 473 Unsupervised ... labelled information for training. Our sys- tem extracts transliteration pairs in an unsupervised fashion. It is also able to utilize labelled information if available, obt...

Ngày tải lên: 19/02/2014, 19:20

... individual models, the model with Brown semantic clusters clearly outper- forms the baseline, but the two models with syntac- tic clusters perform almost the same as the baseline. The ensemble model ... is our ensemble model which is the lin- ear combination of the three cluster-based models. As Table 1 shows, the ensemble model has out- performed the baseline and individual models in...

Ngày tải lên: 17/03/2014, 00:20

... θ c,t [draw sem class for arg] GenSemClass(c  c,t ) [recurse] Figure 2: The generative story for the Bayesian model for unsupervised semantic parsing. tributions over syntactic paths for the argument type ... of the Association for Computational Linguistics, pages 1445–1455, Portland, Oregon, June 19-24, 2011. c 2011 Association for Computational Linguistics A Bayesian Mode...

Ngày tải lên: 23/03/2014, 16:20

... this model, we found 834 entries for the Π vector (which models the distri- bution of tags in first position in sentences) out of possibly N = 1934, about 250K entries for the A matrix (which models ... guidelines we published and checked for agreement. The test corpus contains about 30K words. We compared two unsupervised models over this data set: Word model [W], and Morpheme mod...

Ngày tải lên: 23/03/2014, 18:20

... three lectures is- used for estimating the optimal word block length for representing nodes, the threshold distances for discarding node edges, the number of uniform chunks for estimating tf-idf ... did not try to ad- just our model to optimize its performance on the synthetic data. The smoothing method developed for lecture segmentation may not be appropriate for short segments...

Ngày tải lên: 20/02/2014, 11:21

... higher. Figure 1 demonstrates these phenomena for two leading models, Collins (1999) model 2, a generative model, and Charniak and Johnson (2005), a reranking model. The parser adaptation scenario is ... Experimental Setup We performed experiments with two parsing mod- els, the Collins (1999) generative model number 2 and the Charniak and Johnson (2005) reranking model. For the firs...

Ngày tải lên: 20/02/2014, 12:20

... Association for Computational Linguistics An Approximate Approach for Training Polynomial Kernel SVMs in Linear Time Yu-Chieh Wu Jie-Chi Yang Yue-Shi Lee Dept. of Computer Science and Information ... function (4) for each support vector x i . The situation is even worse when the number of support vectors become huge (Kudo and Matsumoto, 2004). Therefore, whether in training o...

Ngày tải lên: 20/02/2014, 12:20