Book Reviews Tree Adjoining Grammars: Formalisms, Linguistic Analysis and Processing
نویسندگان
چکیده
Take a finite set T of trees, and close it under the operation of substitution—replacing daughterless nonterminals by other trees in T whose root node label matches the nonterminal. The set of all terminal strings of trees in the resultant tree-set will be a context-free language (CFL), and for every CFL there will be such a finite set of trees that generates it under substitution. For an elegant formalization of CFLs in these terms, see Rogers (1999, pages 25–26). Now assume an additional operation: besides substituting trees for daughterless nonterminals, you can also squeeze new material into the middle of a tree, substituting it for a nonterminal node that has daughters. More precisely, assume a finite set A of insertable trees that share a special form: each A ∈ A has a single node on its frontier, known as A's foot node, that has the same label as A's root. Squeezing some A ∈ A into a tree T ∈ T means replacing a node n in T so that n's mother becomes the mother of A's root, and everything dominated by n comes to be dominated by A's foot node. Closing T under the operation of squeezing in new material from A in this manner (an operation called tree adjunction) yields a tree-set of which the set of all terminal strings will be a tree adjoining language (TAL), and for every TAL there will be an appropriate pair T , AA. Such a pair (called a tree adjunct grammar in Joshi, Levy, and Takahashi [1975]) is known today as a tree adjoining grammar (TAG). The research program on TAGs that Aravind Joshi has led since 1975 is perhaps the most interesting and significant research program in formal language theory of the last 40 years. General linguists have clearly underrated it, though computational linguists have in general kept more closely in touch with it. The TALs are a mathematically natural class with closure and decision properties very similar to those of the CFLs, including a polynomial-time recognition problem. Several independent but equivalent characterizations of the class have been discovered: Vijay-shanker and Weir (1994) present a weak equivalence result for head grammars gave a characterization in terms of embedded pushdown automata; and Rogers (1998) gives a new model-theoretic characterization in terms of linearized terminal strings of three-dimensional tree models of monadic second-order logic formulae.
منابع مشابه
The Relationship Between Tree Adjoining Grammars And Head Grammarst
67 We examine the relationship between the two grammatical formalisms: Tree Adjoining Grammars and Head Grammars. We briefly investigate the weak equivalence of the two formalisms. We then turn to a discussion comparing the linguistic expressiveness of the two formalisms.
متن کاملThings between Lexicon and Grammar
A number of grammar formalisms were proposed in 80’s, such as Lexical Functional Grammars, Generalized Phrase Structure Grammars, and Tree Adjoining Grammars. Those formalisms then started to put a stress on lexicon, and were called as lexicalist (or lexicalized) grammars. Representative examples of lexicalist grammars were Head-driven Phrase Structure Grammars (HPSG) and Lexicalized Tree Adjoi...
متن کاملFrom Constraints to TAGs: A Transformation from Constraints into Formal Grammars
We present an approach specifying constraints on labeled trees, such as are commonly used within (computational) linguistic theories, and show how to translate these constraints to formal grammars, in particular context-free grammars and (lexicalized) tree adjoining grammars. Our method offers the formal advantages of constraint-based formalisms and at the same time operational efficiency for t...
متن کاملSynchronous and Multicomponent Tree-Adjoining Grammars: Complexity, Algorithms and Linguistic Applications
This thesis addresses the design of appropriate formalisms and algorithms to be used for natural language processing. This entails a delicate balance between the ability of a formalism to capture the linguistic generalizations required by natural language processing applications and the ability of a natural language processing application based on the formalism to process the formalism efficien...
متن کاملWide-Coverage NLP with Linguistically Expressive Grammars
In recent years, there has been a lot of research on wide-coverage statistical natural language processing with linguistically expressive grammars such as Combinatory Categorial Grammars (CCG), Head-driven Phrase-Structure Grammars (HPSG), Lexical-Functional Grammars (LFG) and Tree-Adjoining Grammars (TAG). But although many young researchers in natural language processing are very well trained...
متن کاملXMG: eXtensible MetaGrammar
In this article, we introduce eXtensible MetaGrammar (XMG), a framework for specifying tree-based grammars such as Feature-Based Lexicalised Tree-Adjoining Grammars (FB-LTAG) and Interaction Grammars (IG). We argue that XMG displays three features which facilitate both grammar writing and a fast prototyping of tree-based grammars. Firstly, XMG is fully declarative. For instance, it permits a de...
متن کامل