Báo cáo khoa học: "Paragraph-, word-, and coherence-based approaches to sentence ranking: A comparison of algorithm and human performance" ppt

... coefficient NoPar agr aph WithParagraph Figure 7. Average rank correlations of algorithm and human sentence rankings. Figure 7 shows average rank correlations (ρ avg ) of each algorithm and human sentence ... simple paragraph-based approach intended as a baseline, two word-based approaches, and two coherence-based approaches. In the paragraph-based approach...

Ngày tải lên: 17/03/2014, 06:20

... Evaluating Answers to Complex Questions: Pyramids Won’t Topple, and Neither Will Human Assessors Hoa Trang Dang Information Access Division National Institute of Standards and Technology Gaithersburg, ... than the single-assessor method of evaluation, the pyramid method also ap- pears to have greater discriminative power. We fit a two-way analysis of variance model with the s...

Ngày tải lên: 20/02/2014, 12:20

... calcu- lated on the basis of constants tabulated by Fabiato and Fabiato [32]. The pCa values were calculated by the computer program EQCAL (Biosoft, Cambridge, UK). Statistics Quantitative values are ... (PIA) were purchased from Molecular Probes. Chromatographic reagents (DEAE-sephadex A- 50 and CM-sephadex C-50) were purchased from Amersham Pharmacia Biotech Asia Pacific. Bicinchonini...

Ngày tải lên: 23/03/2014, 21:20

... [26–28]. This parameter represents the rate of change of the free energy of unfolding as a function of denaturant concentration and is proportional to the amount of additional surface area exposed ... characterization of the mutants in native conditions and in 7.9 M GdmCl In native conditions, the near-UV CD spectra of the three mutants E5 3A, D28 7A and E5 3A ⁄ D28...

Ngày tải lên: 30/03/2014, 04:20

... pitch, amplitude and pronuncia- tion and users are given immediate feedback on the acceptability of each recording. Users can then rerecord an unacceptable utterance. Recordings are automatically ... naturalness and individuality one associates with one’s own voice. Individuals with difficulty speak- ing can be any age, gender, and from any part of the country, with regional...

Ngày tải lên: 20/02/2014, 09:20

... relations having of the order of a thousand tuples. A disadvantage of much current work on NL access to databases is that the work is restricted to providing access to databases, whereas ... computational formalisms. The outcome of this "top-down" approach, as reallsed in the language Prolog, has a great deal in common with the relational approach to databa...

Ngày tải lên: 21/02/2014, 20:20

... the approximation in Equation (1) is to iden- tify all paths containing an n-gram and accumulate their probabilities. The accumulation of probabil- ities at the path level, rather than the n-gram ... Ψ R n , then each path in E n ◦ Ψ R n has only one non-ǫ output la- bel u and all paths leading to a given final state share the same u. A modified forward algorithm can be used to...

Ngày tải lên: 23/03/2014, 16:20

... hydrolyse Ala-Ala-pNA and Ala-pNA (or other chromogenic amino acids) at reason- able rates clearly indicates exclusive cleavage of the anilide bond of Ala-Ala-Val-Ala-pNA. Our results also provide convincing ... the TGase appendage, release of pNA declined by an order of magnitude. The a nity of Pro-Leu-Gly-pNA or Ala-Ala- Phe-pNA for SM-TAP corresponded to that of Ala-Ala- V...

Ngày tải lên: 31/03/2014, 07:20

... src="data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAA9gAAAFXCAIAAADWMXECAAAACXBIWXMAABYlAAAWJQFJUiTwAAAgAElEQVR42uzdeVxTV/438EPCkrAEBJe4AIorErSKCLhCI0WqI0IVFRWwLm1RnHZ0purgr1Z5nNrq6EBLW6tF24KKLYp1wYVCXQBBUEkQ17gAiooQgpCw3OT5485cLwEpVSABPu8Xf5h4c3O4uSSfe/I95xhoNBoCAAAAAADti4NDAAAAAACAIA4AAAAAgCAOAAAAAAAI4gAAAAAACOIAAAAAAIAgDgAAAACAIA4AAAAAAAjiAAAAAAAI4gAAAAAAgCAOAAAAAIAgDgAAAACAIA4AAAAAAAjiAAAA...

Ngày tải lên: 19/02/2014, 19:20

... 2006). They used a natural language tagger which was trained on the output of ParaMor and Morfes- sor. The goal was to mimic each algorithm since ParaMor is rule-based and there is no access to Morfessor’s ... terms of training set size. We want to remind the reader that our two algorithms are aimed at small datasets. We randomly split each dataset into 10 subsets where each...

Ngày tải lên: 20/02/2014, 04:20