Maxwell, J. T. and Kaplan, R. M. (1993). The Interface between Phrasal and Functional Constraints. Computational Linguistics, 19(4):571-- 590.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....English sentence where the subject dogs agrees in number with the verb eat. Features can thus prevent grammars from overgenerating (i.e. generating sentences that are not grammatical) 9 2.3 Generative vs. Eliminative Parsing Natural language parsing can be divided into two approaches [34]. The first is a constructive, or generative, process in which structural descriptions are built out of elementary building blocks (e.g. chart parsing [2] Most commonly known ap proaches to parsing use this technique. The other approach is eliminative parsing. In eliminative parsing, the ....

....Grammar (CDG) which is based on the eliminative parsing approach introduced in the previous chapter. CDG was first described by Maruyama in 1990 [3, 35, 36] and has been subsequently studied in both the United States [4, 21, 102, 103, 104, 105, 106,107,108,109,110,111,112,113,114] and Europe [22, 34, 115,116,117, 118,119, 120, 121]. Like other dependency grammars, CDG describes the syntactic structure of a sentence by how words relate to each other. It can make use of a word s feature and semantic information in order to reduce ambiguity in the parse solutions, and it can also parse mildly context sensitive languages. We ....

[Article contains additional citation context not shown here]

John T. Maxwell III and Ronald M. Kaplan. The interface between phrasal and functional constraints. Computational Linguistics, 19(4):571-590, December 1993.

....backbone, it is possible to take advantage of the efficiency of context free parsing if the parser proceeds in two steps: First a context free parser builds the context free part of the syntactic analysis which is then extended by the calculation of feature structures. Maxwell and Kaplan [Maxwell III and Kaplan, 1994] experimented with variants of this strategy in their LFG parser. One result of their experiments was that their grammar is processed more efficiently by their parser if the rather broad contextfree categories chosen by the grammar writers are replaced by more specific This is a revised version ....

....recomputation of feature structures had a bigger influence on the parsing of the complex sentence than on the parsing of the simpler sentences. The impact of the incorporation of features into the context free grammar has also been examined. We observed in contrast to Maxwell and Kaplan [Maxwell III and Kaplan, 1994], only a marginal speedup of about 3 percent from feature incorporation. The incorporation of some features let to disastrous results because the parse forest generated by the context free parser became very big, slowing down both context free parsing and the calculation of the feature structures. ....

Maxwell III, J. T. and Kaplan, R. M. (1994). The interface between phrasal and functional constraints. Computational Linguistics, 19(4):571--589.

....with real life sentences, sophisticated methods of managing these ambiguities are thus required. 2. 2 Packing ambiguities The parsing and generation algorithms realized in XLE are based on insights from research into ecient processing algorithms for uni cation based grammars (see in particular Maxwell and Kaplan (1989, 1993, 1996) and Shemtov (1997) 1 One important facet of these ecient processing techniques is an algorithm for contexted constraint satisfaction, a method for processing ambiguities eciently in a chart like packed representation. A major source of computational complexity with higher level ....

Maxwell, J. and Kaplan, R. (1993). The interface between phrasal and functional constraints. Computational Linguistics, 19(4):571-590.

.... unification augmented formalisms, parsing is no longer guaranteed to be polynomial time, and in certain cases may even not terminate (when infinitely many f structures can be associated by the grammar with a given substring of the input) 17] Additionally, as pointed out by Maxwell and Kaplan [12], interleaved pruning is not the only possible computational strategy for applying the two types of constraints. However, not only is interleaved pruning the most common approach used, but certain grammars, in which the context free backbone in itself is cyclic, actually rely on interleaved ....

....[20] compares left corner and chart parsers that do apply ambiguity packing with a GLR parser without ambiguity packing. In Proceedings of IWPT 2000, Trento, Italy, February 2000, Pages 147 158 f structures has received quite a bit of attention in the literature over the last decade [4] 11] [12] [13] 6] While the two issues are related, we focus here on how to achieve optimal packing of c structures, so that unification operations do not have to be re executed due to the later detection of an additional local ambiguity. While there is much to be gained from improved packing on the ....

J. T. Maxwell and R. M. Kaplan. The Interface between Phrasal and Functional Constraints. Computational Linguistics, 19(4):571--590, December 1993.

....gap or verb form feature) than just phrase structure category, those features are used to prune unsuccessful rule application at the earliest time. Although with a very small example, a substantial effect on parsing efficiency by the use of of restriction is reported. Another approach taken in (Maxwell and Kaplan, 1994) encodes some (functional) features directly in the context free symbols (which requires the grammar modification) thereby allowing those features to be propagated down by the predictor operation of the Earley s algorithm. Not only does this strategy enable the early detection of parse failure, ....

Maxwell, J. and Kaplan, R. (1994). The interface between phrasal and functional constraints, Computational Linguistics, 19 (4).

....feature of our CFG ltering is bene cial because it eliminates uni cation for partial trees that do not contribute to the nal trees for a whole sentence. The idea to lter out impossible parse trees by a manually tailored CFG backbone within high level grammars has already been explored in lfg (Maxwell and Kaplan, 1995). Our research focus is on the method to automatically extract a CFG backbone from a given hpsg based grammar. In hpsg, a CFG backbone is not given explicitly. The descriptions corresponding to nonterminals in CFGs are given in the form of feature structures, and they interact with each other ....

Maxwell III, J. T., & Kaplan, R. M. (1995). The interface between phrasal and functional constraints. In M. Dalrymple, R. M. Kaplan, J. T. Maxwell III, & A. Zaenen (Eds.), Formal issues in Lexical-Functional Grammar (pp. 403 - 429). Stanford, CA: CSLI Publications.

....f structure is a representation of the functional structure of a sentence, using categories such as Subject, Object, and so on. C and f structure are related by functional constraints associated with CFG rules, called functional schemata. As has been discussed in the LFG and related literature (Maxwell and Kaplan, 1993), parsing grammars that are associated with functional constraints is computationally costly, the run time being exponential in the length of the input string in the worst case. Burheim (1996) proposes to replace the CFG based characterization of c structure with a tree adjoining grammar (TAG) ....

....a linguistically motivated analysis (word order in West Germanic, wh movement in non wh initial languages such as Kashmiri) functional uncertainty also covers such cases. This paper proposes, for the first time, a TAG related framework that directly models functional uncertainty. 3 2 Also see (Maxwell and Kaplan, 1993) for a discussion of how modifying the c structure CFG in a less radical manner can also lead to improvements in parsing. 3 Burheim (1996) mentions a possible treatment of functional uncertainty, but the approach is not worked out in any detail. 2 2 UVG RegDL Rambow (1994) introduces a new ....

Maxwell, J. T. and Kaplan, R. M. (1993). The interface between phrasal and functional constraints. Computational Intelligence, 19(4):571--590.

.... tools (Kuhn et al. 1998) Algorithms and architectures for unificationbased grammar processing The parsing and generation algorithms realized in XLE are based on insights from research into efficient processing algorithms for parsing and generation with unification based grammars, in particular (Maxwell and Kaplan 1989, 1993, 1996) and (Shemtov 1997) While context free phrase structure grammars allow for parsing in polynomial time, grammar formalisms that in addition specify feature constraints can be NP complete or undecidable, and parse in worst case exponential or infinite time. Maxwell and Kaplan (1993) ....

....(Maxwell and Kaplan 1989, 1993, 1996) and (Shemtov 1997) While context free phrase structure grammars allow for parsing in polynomial time, grammar formalisms that in addition specify feature constraints can be NP complete or undecidable, and parse in worst case exponential or infinite time. Maxwell and Kaplan (1993) investigate the computational properties of standard hybrid parsing architectures for unification based grammars. They propose alternative non interleaved processing architectures, which exploit various computational interface properties that support sub exponential parsing. The unification ....

[Article contains additional citation context not shown here]

Maxwell III, J.T. and Kaplan, R.M. 1993. The interface between phrasal and functional constraints. In Computational Linguistics, 19(4):571--590.

....The XLE platform integrates an eOEcient parser and generator for LFG grammars. The parsing and generation algorithms are based on insights from research into eOEcient processing algorithms for parsing and generation with unication based grammars, in particular (Maxwell and Kaplan 1989) (Maxwell and Kaplan 1993), Maxwell and Kaplan 1996) and (Shemtov 1997) While context free phrase structure grammars allow for parsing in polynomial time, grammar formalisms that in addition specify feature constraints can be NP complete or undecidable, and parse in worst case exponential or innite time. However, the ....

Maxwell, J. T. and R. M. Kaplan (1993). The interface between phrasal and functional constraints. Computational Linguistics 19 (4), 571590.

....into high level parameter values , which tasks may be optional and or multiple and under which conditions, and which pre and post conditions must hold for a subtask execution. In this sense, tasks and rules in this system act as Unification Grammars in Natural Language Processing systems [9]. ATOMS task models are specified using a declarative modeling language which incorporates inheritance both for tasks and rule definitions, in order to improve the reuse of partial models. Applying this inheritance concept, a task can be seen as a prototype for other task instances, which may be ....

Maxwell, J.T. and Kaplan, R. M. "The Interface between Phrasal and Functional Constraints", Computational Linguistics, no. 4, 1994.

....uses an identier, followed by one or more parameters in square brackets (e.g. ZP[f,g] The original intention was to provide a mechanism for transforming certain feature distinctions into c structure distinctions (the processing advantages of doing so will be discussed in sec. 4. 3, cf. also Maxwell and Kaplan 1995). Thus, a simple example for a rule specication using complex categories (ignoring f annotations) is the following: PP[var] P NP[var] NP[var] Ppost . The NP and VP categories in this example are viewed as further subclassied, thus their category symbol is complex. Rule specication ....

....generality; this means that at least analyses inspired by extended head theory can be integrated in existing larger grammars. 4. 3 The reAEex of Endocentricity and Fully extended coherence on c structure The key idea for avoiding the observed complexity problem is quite simple and not new (cf. Maxwell and Kaplan 1995): additional information is included at the (technical) level of c structure, such that ill formed analyses are excluded during c structure (context free) parsing already reducing the search space for the later parsing steps. 14 It is important to note the dioeerence between the view of ....

Maxwell, J. and R. Kaplan (1995). The interface between phrasal and functional constraints. In M. Dalrymple, R. Kaplan, J. Maxwell, and A. Zaenen (Eds.), Formal Issues in Lexical-Functional Grammar. Stanford, CA: CSLI Publications.

.... Grammar [25] Generalized Phrase Structure Grammar [26] Unification Categorial Grammar [27, 28] Head Driven Phrase Structure Grammar [29] Definite Clause Grammars [30] and Lexicalized Tree Adjoining Grammar [31] make use of lexical and semantic properties by using feature structures [32, 33]. As Johnson [32] points out, these feature structures are specified indirectly in terms of constraints that they must satisfy. For example, in English a verb and its subject must agree in number. Thus, while both of the following sentences: the dog eat the bones and the dogs eat the bones ....

John T. Maxwell III and Ronald M. Kaplan. The interface between phrasal and functional constraints. Computational Linguistics, 19(4):571--590, December 1993.

....the constraints expressed in the context free backbone into feature constraints and solves the constraint system as a whole. An advantage of the last strategy is the uniformity of the processing mechanism and the ability to process grammars without a context free backbone. Maxwell and Kaplan [Maxwell III and Kaplan, 1994] explore several variants of the first two strategies in their LFG parser. They conclude : that non interleaved pruning is always better than interleaved pruning. The reason probably is that the time spent on context free parsing is neglectable compared to the time spent on the feature ....

....unifications. Avoiding unnecessary recomputation of feature structures had a larger influence on the parsing of the complex sentence than on the parsing of the simpler sentences. The impact of the incorporation of features into the context free grammar was examined, as well. In contrast to [Maxwell III and Kaplan, 1994], I only observed a small speedup of 3 percent for the best combination of incorporated features compared to parsing without feature incorporation. Incorporation of morpho syntactic features like number, gender and case made the parser very slow because the parse forest generated by the ....

Maxwell III, J. T. and Kaplan, R. M. (1994). The interface between phrasal and functional constraints. Computational Linguistics, 19(4):571--589.

....real life sentences, sophisticated methods of managing these ambiguities are thus required. 2. 2 Packing ambiguities The parsing and generation algorithms realized in XLE are based on insights from research into efficient processing algorithms for unification based grammars (see in particular Maxwell and Kaplan (1989, 1993, 1996) and Shemtov (1997) 4 One important facet of these efficient processing techniques is an algorithm for contexted constraint satisfaction, a method for processing ambiguities efficiently in a chart like packed representation. A major source of computational complexity with ....

Maxwell, J. and Kaplan, R. (1993). The interface between phrasal and functional constraints. Computational Linguistics, 19(4):571--590.

No context found.

Maxwell, J. T. and Kaplan, R. M. (1993). The Interface between Phrasal and Functional Constraints. Computational Linguistics, 19(4):571-- 590.

No context found.

John T. Maxwell III and Ronald M. Kaplan. 1993. The interface between phrasal and functional constraints. Computational Linguistics, 19(4):571 589.

No context found.

John Maxwell and Ron Kaplan. 1993. The interface between phrasal and functional constraints. Computational Linguistics, 19(4):571--589.

No context found.

John Maxwell and Ronald Kaplan. 1993. The interface between phrasal and functional constraints. Computational Linguistics, 19(4):571--589.

No context found.

Maxwell, J. T. and R. M. Kaplan, The interface between phrasal and functional constraints, Computational Intelligence 19 (1993), pp. 571--590.

No context found.

MAXWELL III J. T., KAPLAN R. M., "The interface between phrasal and functional constraints", Computational Linguistics, vol. 19, num. 4, 1993, p. 571--590.

No context found.

John T. Maxwell and Ronald M. Kaplan. The interface between phrasal and functional constraints. Computational Linguistics, 19(4):571--590, December 1993.

No context found.

John T. Maxwell and Ronald M. Kaplan. 1993. The interface between phrasal and functional constraints. Computational Linguistics, 19(4):571--590, December.

No context found.

John T. Maxwell and Ronald M. Kaplan. 1993. The interface between phrasal and functional constraints. Computational Linguistics, 19(4):571--590, December.

No context found.

John T. Maxwell and Ronald M. Kaplan. The interface between phrasal and functional constraints. Computational Linguistics, 19(4):571--590, December 1993.

No context found.

J. T. Maxwell III and R. M. Kaplan. The Interface between Phrasal and Functional Constraints. Computational Linguistics, 19(4):571--590, 1993.

First 50 documents

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