Báo cáo khoa học: "Using aggregation for selecting content when generating referring expressions" pdf

... Using aggregation for selecting content when generating referring expressions John A. Bateman Sprach- und Literaturwissenschaften ... information has of course no for- mal consequences for the information contained in the lattice. To determine RE -content appropriate for re- ferring to object ol, we again look for minimal (i.e., ... 'confuse' the objects...

Ngày tải lên: 08/03/2014, 06:20

... provides much useful information for subsequent process- ing. However, this is generally at the cost of re- duced coverage, due both to the difficulty of pro- viding analyses for all phenomena, and ... practice this is not always true. For example, to improve performance, semantically empty lexemes are not inserted into a generation unless a “trigger-rule” defines a context for them. Th...

Ngày tải lên: 31/03/2014, 00:20

... the Association for Computational Linguistics:shortpapers, pages 1–5, Portland, Oregon, June 19-24, 2011. c 2011 Association for Computational Linguistics Lexicographic Semirings for Exact Automata ... and motivate its use for speech and language processing tasks. We prove that the semiring allows for exact en- coding of backoff models with epsilon tran- sitions. This allows for...

Ngày tải lên: 20/02/2014, 04:20

... performance end end average_performance and keep average_performance end B = best average_performance bestFeatures  B ∪ bestFeatures end Figure 2. Implemented algorithm for forward ... saved for later experiments. 4.1 Selecting the features The first task was to reduce the dimensionality and find the best set of features for maximizing the PPV for Emotional c...

Ngày tải lên: 20/02/2014, 16:20

... ∅; ii) For each sense s ∈ V , and for each lexico- semantic relation in D connecting sense s to s  ∈ V , we perform: E := E ∪ {(s, s  )}; iii) For each sense s ∈ V , let gloss(s) be the set of content ... tagged gloss. Then for each content word w  in gloss(s) and for each sense s  of w  , we add the corresponding edge to the dictionary graph, i.e.: E := E ∪ {(s, s  )}. F...

Ngày tải lên: 22/02/2014, 02:20

... transform syntactic as well as morphological variations of names into a normalized name form. Basic trans- formations can be achieved via pattern match- ing (regular expressions) while for more ... and other information like the numbering of atoms. Thus, no additional information has to be con- sidered to generate a chemical representation for- mat from the graph. We focus on generating...

Ngày tải lên: 08/03/2014, 01:20

... tags directly from lexical information, which is eas- ily scalable for languages that lack suf- ficient parsing resources or have inher- ent linguistic challenges for parsing. We investigated ... satisfactory for practical use. The limitation is even more pertinent for the lan- guages that do not have sophisticated parsing re- sources, or languages that have inherent linguistic challen...

Ngày tải lên: 08/03/2014, 01:20

... the Association for Computational Linguistics, pages 103–108, Jeju, Republic of Korea, 8-14 July 2012. c 2012 Association for Computational Linguistics INPRO_iSS: A Component for Just-In-Time ... Just-In-Time Incremental Speech Synthesis Timo Baumann University of Hamburg Department for Informatics Germany baumann@informatik.uni-hamburg.de David Schlangen University of Bielefeld Facu...

Ngày tải lên: 16/03/2014, 20:20

... readied with information from the “outside”—the input information. Corresponding to each input state will be an output state, signaled by a “stop” or some such indicator. When the information from ... formation. The output state is a function of the input state, and correspondingly, the output information is a function of the input information. An automaton, in its capacity as a mean...

Ngày tải lên: 23/03/2014, 13:20

... w j =before(w i ). It can be applied recursively, as w k = before(w j ) = before(before(w i )) to generate the entire sentence starting with an arbitrary word, usually the sentence Main. before() ... applied as a predicate, such as before(w i , w j ). Thus for each word w i in the sentence, inSentence(w i , s i ) ⇒ main(w m , s k ) ∧ (before(w i , w m ) ∨ before(w m , w i )). A con...

Ngày tải lên: 23/03/2014, 19:20