Yves Schabes. Stochastic lexicalized tree-adjoining grammars. In Proceedings of the 14 International Conference on Computational Linguistics (COLING'92), 1992.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....4.3 Formalization of Probabillstlc TAG There are a number of ways to express lexicalized tree axljoining grainmar as a probabilistic grammar formalism. 4 Here I propose what appears to me to be the most direct probabilistic generalization of lexicalized TAG; a different treatment can be found in [Schabes, 1992]. Definitions: Let I denote the set of initial trees in the grammar, and A the set of auxiliary trees. Each tree in a lexicalized TAG has a (possibly empty) subset of its frontier nodes marked as nodes at which substitution may take place. Given a tree a, let that subset be denoted by ....

....by the trees that they anchor, regardless of iutervening material in the surface string. 4.4 Acquisition of Probabilistic TAG Despite the attractive theoretical features of probabilistic TAG, many practical issues remain. Foremost among these is the acquisition of the statistical parameters. [Schabes, 1992] has recently adapted the InsideOutside algorithm, used for estimating the parameters of a probabilistic CFG [Baker, 1979] to probabilistic TAG. The InsideOutside algorithm is itself a generalization of the Forward Backward algorithm used to train bidden Markov models [Baum, 1972] It optimizes ....

Yves Schabes. Stochastic lexicalized treeadjoining grammars, 1992. This proceedings.

....them. Examples of such tasks include searching a word graph for the most probable sentence (MPS) according to a Stochastic Context Free Grammar (SCFG) 11] or finding, for a given sentence, the most probable parse tree (MPP) according to a Stochastic Lexicalized Tree Adjoining Grammar (SLTAG) [16] or a Data Oriented Parsing (DOP) model [5] The problem with such maximisation tasks is that they often correspond to an instance of an NP hard problem [17] and, as such, cannot be solved exactly in an e#ective way. In such cases, heuristics and or approximations need to be used instead. ....

Yves Schabes. Stochastic lexicalized tree-adjoining grammars. In Proc. 14th Int. Conf. of Computationnal Linguistics (COLING), pages 426--432, Nantes (France), August 1992.

.... the EM algorithm for the unsupervised training of Probabilistic Context Free Grammars known as the Inside Outside algorithm has been found in practice to be computationally intractable for realistic problems [1] Unsupervised learning algorithms have been designed for other grammar models (e.g. [2, 3]) However, to the best of our knowledge, no large scale experiments have been carried out to test the ecacy of these algorithms; the most likely reason is that their computational complexity, like that of the Inside Outside algorithm, is impractical. One way to improve the complexity of ....

Yves Schabes. Stochastic lexicalized tree-adjoining grammars. In Proceedings of the Fourteenth International Conference on Computational Linguistics, pages 426-432, Nantes, France, 1992.

....modeling is the task of finding a distribution over strings, a task made feasible by computers and originally of interest to computer scientists and engineers working on speech recognition. It relies on fairly simple statistical models which require large amounts of data and computation (e.g. [3, 4]) The former approach ignores statistical methods in favor of expert knowledge, while the latter commits the inverse error. It seems clear that both sides could benefit by borrowing ideas from the other. In this research, we develop methods for learning probabilistic context free grammars ....

....seen It seems likely that the probabilities for the grammar must be revised in such a way as to reduce the probability of some sentences that were previously parsable. In that case, 17 however, the reported probabilities do not make sense, as they may sum to more than 1.0. Like Hindle, Schabes [4,23] reports work on lexicalized grammars, in this case lexicalized tree adjoining grammars. As with Hindle, he faces a sparse data problem, as each word may be associated with several syntactic rules. His tack on this problem is to use more informative data. Rather than training his grammar with raw ....

Yves Schabes, "Stochastic Lexicalized Tree-Adjoining Grammars," Proc. of COLING-92 (1992).

.... of class based language models , we plan to consider additional distributional relations ( for instance , adjective noun ) and apply the results of clustering to the grouping of lexical associations in lexicalized grammar frameworks such as stochastic lexicalized tree adjoining grammars REF Schabes 1992 REF . S P DIV DIV DEPTH= 1 HEADER ID= H 15 Acknowledgments HEADER P S ID= S 169 We would like to thank Don Hindle for making available the 1988 Associated Press verb object data set , the Fidditch parser and a verb object structure filter , Mats Rooth for selecting the ....

....et al. 1990 REFLABEL Kenneth SURNAME Rose SURNAME , Eitan SURNAME Gurewitz SURNAME , and Geoffrey C. SURNAME Fox SURNAME . DATE 1990 DATE . Statistical mechanics and phase transitions in clustering. Physical Review Letters, 65(8) 945 948. REFERENCE REFERENCE REFLABEL Schabes 1992 REFLABEL Yves SURNAME Schabes SURNAME . DATE 1992 DATE . Stochastic lexicalized tree adjoining grammars. In Proceeedings of the 14th International Conference on Computational Linguistics, Nantes, France. REFERENCE REFERENCE REFLABEL Yarowsky 1992 REFLABEL David ....

[Article contains additional citation context not shown here]

Yves Schabes. 1992. Stochastic lexicalized tree-adjoining grammars. In Proceeedings of the 14th International Conference on Computational Linguistics, Nantes, France.

....them. Examples of such tasks include searching a word graph for the most probable sentence (MPS) according to a Stochastic Context Free Grammar (SCFG) 11] or finding, for a given sentence, the most probable parse tree (MPP) according to a Stochastic Lexicalized Tree Adjoining Grammar (SLTAG) [16] or a Data Oriented Parsing (DOP) model [5] The problem with such maximisation tasks is that they often correspond to an instance of an NP hard problem [17] and, as such, cannot be solved exactly in an effective way. In such cases, heuristics and or approximations need to be used instead. ....

Yves Schabes. Stochastic lexicalized tree-adjoining grammars. In Proc. 14th Int. Conf. of Computationnal Linguistics (COLING), pages 426--432, Nantes (France), August 1992.

.... Language Processing and More HASIDA Koiti Natural Language Section, Electrotechnical Laboratory 1 Introduction Natural language processing (NLP) has employed statistical tools such as Hidden Markov Model (HMM) 4] Stochastic Context Free Grammar (SCFG) Stochastic Tree Adjoining Grammar (STAG) 2] Probabilistic Generalized LR Parser (PGLRP) 3] etc. Below we propose a general method which subsumes all of them. In this method, constraint is represented as a Horn logic program whose semantics is a probability distribution over candidates for solutions. Symbolic computation is to transform ....

....distribution over candidates for solutions. Symbolic computation is to transform this program so as to figure out high probability solutions while discarding wrong hypotheses. Structure sharing in the program guarantees efficient computation for both symbolic and probabilistic information. 2 Probabilistic Constraint A constraint is represented by a Horn program such as in Figure 1. Names beginning with capital (a) s(A0,z) A0=a(A1) A1=b(A2) A2=a(z) b) s(B,C) B=a(C) c) s(D,E) s(D,F) t(F,E) d) s(G,H) u(G,I) s(I,H) e) t(J,K) J=b(L) s(L,K) f) u(M,N) ....

[Article contains additional citation context not shown here]

Yves Shabes. Stochastic lexicalized tree-adjoining grammars. In Christian Boitet, editor, Proceedings of the Fourteenth International Conference on Computational Linguistics, pages 426-- 432, Nantes, 1992.

No context found.

Yves Schabes. Stochastic lexicalized tree-adjoining grammars. In Proceedings of the 14 International Conference on Computational Linguistics (COLING'92), 1992.

No context found.

Yves Schabes. Stochastic lexicalized tree-adjoining grammars. In Proceedings of the Fourteenth International Conference on Computational Linguistics (COLING-92), pages 426--432, Nantes, 1992.

No context found.

Yves Schabes. Stochastic lexicalized tree-adjoining grammars. In Proceedings of the Fourteenth International Conference on Computational Linguistics, pages 426--432, Nantes, France, 1992.

No context found.

Shabes, Y. (1992). Stochastic lexicalized tree-adjoining grammars. In Proceeding of the 14 International Conference on Computational Linguistics.

No context found.

Yves Schabes. Stochastic lexicalized tree-adjoining grammars. In Proceedings COLING'92, Nantes, 1992.

No context found.

Yves Schabes. Stochastic lexicalized tree-adjoining grammars. In Proceedings of the Fourteenth International Conferenceon Computational Linguistics #COLING '92#, Nantes, France, July 1992.

No context found.

Yves Schabes. 1992. Stochastic lexicalized treeadjoining grammars. In Proc. 14th COLING, pages 426--432.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC