Báo cáo khoa học: "A Flexible Stand-Off Data Model with Query Language for Multi-Level Annotation" ppt

... input: • A database with facts about the selected car (maximum speed, horse power, etc.). • A database which correlates facts with value judge- ments. 1 NECA stands for 'Net Environment for Embodied ... A Flexible Pragmatics-driven Language Generator for Animated Agents Paul Piwek ITRI — Information Technology Research Institute University of ... NATURAL LANGUAGE GENERA-...

Ngày tải lên: 22/02/2014, 02:20

... photoperiod at 22 °C with cool white light (120 mmolÆphotonsÆm )2 Æs )1 ), and used for trans- formation. For different stresses, 2-week-old seedlings were transferred to blotting paper without stress ... AtHELPS- overexpressing lines, Col-0 plants were transformed with a 35S::AtHELPS construct. Homozygous trans- formant seedlings were screened with kanamycin selec- tion, increased...

Ngày tải lên: 14/02/2014, 18:20

... fatty acids or their metabolites are agonists for PPAR-d or antagonists for PPAR-c, and this is the molecular basis for cancer formation, at least in some model systems. These theories merit further ... acids, KASL [49]. These studies were performed before the existence of the minor splice vari- ants [81–83] was known, and therefore refer to AMA- CR 1A, the major form of the enzyme in...

Ngày tải lên: 18/02/2014, 17:20

... and the GPX activity for 6-CySeCD was higher than for 6-SeCD. Kinetics of 6-CySeCD Steady-state kinetics was observed for substrates H 2 O 2 and GSH. The initial velocities for reduction of H 2 O 2 by ... (b-CD) cavity provided a hydrophobic environ- ment for substrate binding [19–23]. For example, 6,6’-diseleno-bis-cyclodextrin (6-SeCD) activity for the reduction of hydrogen pe...

Ngày tải lên: 19/02/2014, 02:20

... and AFNLP A Syntax-Driven Bracketing Model for Phrase-Based Translation Deyi Xiong, Min Zhang, Aiti Aw and Haizhou Li Human Language Technology Institute for Infocomm Research 1 Fusionopolis ... bracketing instances for each index j: 1) the bracketable instance with the minimal τ(c i k ), 2) the bracketable instance with the maximal τ(c i k ), 3) the unbracketable instance with...

Ngày tải lên: 20/02/2014, 07:20

... transform model works in a dif- ferent way, by building up constituent structure from left to right. In Figure 2, the same fragment is shown as it appears in the training data for this sys- tem. With ... Right-corner transform Binarized trees 2 are then transformed into right- corner trees using transform rules similar to those described by Johnson(1998a). This right-corner transform...

Ngày tải lên: 20/02/2014, 09:20

... to 0.1 M with buffer B, fraction V was loaded onto a ssDNA-cellulose column equilibrated with buffer B. After washing the column excessively with buffer B the bound proteins were eluted in steps with ... activity upon heating at 56 °C for 1 min (data not shown). Significant unwinding activity was observed in the broad pH range (pH 7.5–9.0) with an optimum near pH 8.0 (data not sh...

Ngày tải lên: 20/02/2014, 11:20

... normalization model consists of two sub-models: a word-based language model (LM), characterized by 1 (| ) nn P ee − ) k and a phrase- based lexical mapping model (channel model) , characterized ... updated model a phrase-based nor- malization model. normalization with a total of 2300 transforma- tions. Substitution accounts for almost 86% of all transformations. Deletio...

Ngày tải lên: 20/02/2014, 12:20

... reagents. Sequence analysis Searches for homologous proteins on GenBank databases (National Center for Biotechnology Information, USA) were performed using the BLAST program, with deduced amino acid sequences ... the synthetic toxin (sWntx-5) agree with those anticipated for the natural toxin. We show that this toxin interacts with relatively low affinity (K d ¼ 180 n M ) with the...

Ngày tải lên: 21/02/2014, 03:20

... cannot afford the computa- tional expense of inducing or applying a full translation model. For these applications, we have designed a fast algorithm for esti- mating a partial translation model, ... lexicon that our model induced from this tiny bitext accounted for 30% of the word types with precision between 84% and 90%. Recall drops when there is tess training data, be...

Ngày tải lên: 22/02/2014, 03:20