Tài liệu Báo cáo khoa học: "A Finite-Slate Parser for Use in Speech Recognition" pdf

... allowable grouping of information, thus expressing the information in- terdependencies. The language which SUG pro- vides for specifying these descriptions allows par- tiality both in the information ... under- specify ordering constraints between nodes, thus allowing for variations in word order. This partial- ity in structural information is imperative to allow incrementa...

Ngày tải lên: 20/02/2014, 21:20

... language or manually annotating a body of text to be used as a training set for a statistical engine or machine learning. In this paper, we focus on using the multilingual Wikipedia (wikipedia.org) ... which contains little content but typically lists a number of entries which might be what the user was seeking. For instance, the page “Franklin” contains 70 links, including the...

Ngày tải lên: 20/02/2014, 09:20

... differentiation was initiated at 48 h post-transfection by maintaining the cells in culture medium containing 2% horse serum. Cells were fixed on day 4 of differentiation and stained with hematoxylin and eosin. ... described in Fig. 1. For hypoxia treatment, 3T3-L1 cells were placed in a hypoxia incubator with 1% O 2 immediately after addi- tion of the IDM cocktail. The control was mai...

Ngày tải lên: 18/02/2014, 08:20

... and incubated in blocking serum containing anti-mouse IgG (Vectastain) for a fur- ther 30 min. Staining was developed by incubation of cells with Vectastain ABC reagent for 30 min, followed by incubation ... Staining was performed according to the manufacturer’s instructions. Briefly, endogenous peroxidases were blocked in 0.3% H 2 O 2 , rinsed in NaCl ⁄ P i and incubated in blocki...

Ngày tải lên: 18/02/2014, 13:20

... and immunoblotting Polyclonal antibodies to War1p were raised in rabbits against a GST-War1p fusion protein containing 294 amino acids of the C-terminal part of War1p (amino acids 650– 944) fused in- frame ... consequence of impaired dimer for- mation of War1p or a lack of DNA binding. Indeed, in the case of War1-42p, in vivo footprinting data (Fig. 5) indicate an inability to bind t...

Ngày tải lên: 19/02/2014, 00:20

... labelled information for training. Our sys- tem extracts transliteration pairs in an unsupervised fashion. It is also able to utilize labelled information if available, obtaining improved performance. We ... results. 1 Introduction Transliteration mining is the extraction of translit- eration pairs from unlabelled data. Most transliter- ation mining systems are built using labelled train-...

Ngày tải lên: 19/02/2014, 19:20

... 3). According to the observation made by Lu and Wolfe [23], who used a combination strategy of 4,6-diamino-2-phenylindole (DAPI) staining for the detection of DNA and Azo dye staining for the identification ... following experiments were car- ried out in the following conditions: 200 lL reaction containing 20 lg protein, incubated for 120 min at 30 °C. The putative DNase had an optimum...

Ngày tải lên: 20/02/2014, 03:20

... | 3 |e| 3 )) that inference is impractical for real translation data. Proposed solutions to this problem include imposing sentence length limits, using small train- ing corpora and constraining the search ... non-terminal and terminal expansions. The former rewrites a non- terminal symbol as a string of two or three non- terminals along with an alignment, specifying the corresponding orderi...

Ngày tải lên: 20/02/2014, 07:20

... combination. 2 Training/Testing The training and testing phases slightly differ from each other. In the training phase, the pairs that are to be used as training examples have to be selected in a process ... merging these information into markables that are the start- ing point for the mentions used by the coreference resolution proper. Starting out with a chunking pipeline, which uses...

Ngày tải lên: 20/02/2014, 09:20

... are computed and search is called again. One interesting property of this framework is that it allows that use of future information in ad- dition to past information. The pipeline model nat- urally allows ... focus point. Next we describe several policies for determin- ing the focus point of the algorithm following an action. We note that, with a few exceptions, de- termining the focus...

Ngày tải lên: 20/02/2014, 12:20