Báo cáo khoa học: "Combining Trigram-based and Feature-based Methods for Context-Sensitive Spelling Correction" docx

Báo cáo khoa học: "Exploring Distributional Similarity Based Models for Query Spelling Correction" docx

Báo cáo khoa học: "Exploring Distributional Similarity Based Models for Query Spelling Correction" docx

... designing a spelling correction program for web search queries, however, poses special technical challenges and cannot be well solved by general purpose spelling correction methods. Cucerzan and ... probabilities from corpus (Church and Gale, 1991; Mayes et al, 1991; Ris- tad and Yianilos, 1997; Brill and Moore, 2000; and Ahmad and Kondrak, 2005). In the men- tioned...

Ngày tải lên: 17/03/2014, 04:20

8 309 0
Tài liệu Báo cáo khoa học: "Combining Functionality and Object Orientedness for Natural Language Processing" ppt

Tài liệu Báo cáo khoa học: "Combining Functionality and Object Orientedness for Natural Language Processing" ppt

... descendants, say z and y, a PAL expression for the phrase is one of the following forms: <z>( <V>) or <p>( <z>) where ~a> stands for a PAL expression for a phrase ~*. ... composite expressions are constructed only with a binary form of function application. Thus, if z and I/ are well-formed formulas of PAL, so is a form z(y). Expressions of PAL ar...

Ngày tải lên: 21/02/2014, 20:20

4 423 0
Báo cáo khoa học: "Combining Trigram and Winnow in Thai OCR Error Correction" potx

Báo cáo khoa học: "Combining Trigram and Winnow in Thai OCR Error Correction" potx

... areas. Make hypotheses for nonwords and unknown words: (a) For each dubious string obtained from 1., the surrounding words are also con- sidered to form candidates for correc- tion by concatenating ... string. For example, in "in- form at j off', j is an unknown string representing a dubious area, and in- form at and on are words. In this 838 case, the u...

Ngày tải lên: 23/03/2014, 19:20

7 385 0
Báo cáo khoa học: "Combining Distributional and Morphological Information for Part of Speech Induction" doc

Báo cáo khoa học: "Combining Distributional and Morphological Information for Part of Speech Induction" doc

... DM uses morphological informa- tion as well, DF uses frequency information and DMF uses morphological and frequency informa- tion. We evaluated it for all words, and also for words with frequency ... range from 12.1 for English to 4.84 for Hungarian. The tags used are extremely fine-grained, and incorporate a great deal of infor- mation about case, gender and so on — in Hun-...

Ngày tải lên: 31/03/2014, 20:20

8 275 0
Báo cáo khoa học: "Combining Statistical and Knowledge-based Spoken Language Understanding in Conditional Models" pptx

Báo cáo khoa học: "Combining Statistical and Knowledge-based Spoken Language Understanding in Conditional Models" pptx

... leaving no hidden variables and resulting in a CRF. Here, PAC stands for “preamble for arrival city,” and PDC for “preamble for departure city.” The command prior and state transition features ... context and chunk coverage features. The chunk coverage feature has three settings: 0 stands for no chunk coverage features; 1 for chunk coverage features for preamble w...

Ngày tải lên: 08/03/2014, 02:21

8 524 0
Báo cáo khoa học: "Combining Acoustic and Pragmatic Features to Predict Recognition Performance in Spoken Dialogue Systems" pdf

Báo cáo khoa học: "Combining Acoustic and Pragmatic Features to Predict Recognition Performance in Spoken Dialogue Systems" pdf

... GUI. The DMT and AT are the core components of In- formation States. The SL and MB are subsidiary data-structures needed for interpreting and generat- ing anaphoric expressions and definite NPs. ... of Informa- tion States (IS) and the update procedures for pro- cessing user input and generating system responses. Here, we briefly introduce parts of the IS which are needed to und...

Ngày tải lên: 08/03/2014, 04:22

8 314 0
Báo cáo khoa học: "Combining data and mathematical models of language change" ppt

Báo cáo khoa học: "Combining data and mathematical models of language change" ppt

... the V forms of {1,1} pairs to be mis- perceived as σ´σ, and for the N forms of {2,2} pairs to be misperceived as ´σσ. 3 Modeling preliminaries We first describe assumptions and notation for models ... both — is possible for some parameter values. (0, 0) and (1, 1) (corresponding to {1,1} and {2,2}) are technically never possible, but effectively occur for FPs of the form (κ, 0)...

Ngày tải lên: 17/03/2014, 00:20

11 407 0
Báo cáo khoa học: "Combining Deep and Shallow Approaches in Parsing German" pptx

Báo cáo khoa học: "Combining Deep and Shallow Approaches in Parsing German" pptx

... dependent in left-to-right or- der (e.g. 0 for in , 1 for on in example (5)), and a number designating the head in left-to-right (e.g. 0 for saw , 1 for man , 2 for hill in (5)). If the links are stored ... representations, Riezler et al. (2002) distinguish upper and lower bound, standing for optimal performance in disam- biguation and average performance, respectively. In I...

Ngày tải lên: 17/03/2014, 06:20

8 417 0
Báo cáo khoa học: "Combining Linguistic and Gaze Features to Resolve Second-Person References in Dialogue" docx

Báo cáo khoa học: "Combining Linguistic and Gaze Features to Resolve Second-Person References in Dialogue" docx

... GDP as the candidate. Hence, we compute a candidate target for the utterance overall, for each third of the ut- terance, and for the period -/+ 2 seconds from the 276 you start time, and in addition, ... http://corpus. amiproject.org/documentations/guidelines-1/ 275 For each participant Pi – target for whole utterance – target for first third of utterance – target for second...

Ngày tải lên: 17/03/2014, 22:20

9 242 0
w