Showing papers on "Context-sensitive grammar published in 1987"

The Formal Complexity of Natural Language

Abstract: Prologue.- What is Mathematical Linguistics?.- I. Early Nontransformational Grammar.- to Part I.- Formal Linguistics and Formal Logic.- An Elementary Proof of the Peters-Ritchie Theorem.- On Constraining the Class of Transformational Languages.- Generative Grammars without Transformation Rules-A Defense of Phrase Structure.- A Program for Syntax.- II Modern Context-Free-Like Models.- to Part II.- Natural Languages and Context-Free Languages.- Unbounded Dependency and Coordinate Structure.- On Some Formal Properties of MetaRules.- Some Generalizations of Categorial Grammars.- III More than Context-Free and Less than Transformational Grammar.- to Part III.- Cross-serial Dependencies in Dutch.- Evidence Against the Context-Freeness of Natural Language.- English is not a Context-Free Language.- The Complexity of the Vocabulary of Bambara.- Context-Sensitive Grammar and Natural Language Syntax.- How Non-Context Free is Variable Binding?.- Prologue.- Computationally Relevant Properties of Natural Languages and Their Grammars.- Index of Languages.- Name Index.

Functional Unification Grammar Revisited

Abstract: In this paper, we show that one benefit of FUG, the ability to state global constraints on choice separately from syntactic rules, is difficult in generation systems based on augmented context free grammars (e.g., Definite Clause Grammars). They require that such constraints be expressed locally as part of syntactic rules and therefore, duplicated in the grammar. Finally, we discuss a reimplementation of FUG that achieves the similar levels of efficiency as Rubinoff's adaptation of MUMBLE, a deterministic language generator.

Reference grammars and pedagogical grammars

Abstract: Four types of grammar are distinguished according to mode of use: reference grammars, pedagogical grammars, theoretical grammars, and teach-yourself grammars. Four factors are shown to enter into a typology of grammars: the mode of use, the language of the user, the level of the user, and the aims of use. Reference grammars and pedagogical grammars are characterized, the former with particular respect to A Comprehensive Grammar of the English Language. In general it is not desirable to attempt to combine in one book the functions of these two types of grammars.

On array grammars and languages

Abstract: Generalized Context-Free (Regular) Kolam Array Grammars [GCF(R)KAG] are introduced as models for generation of rectangular arrays. These grammars are found to be richer in generative capacity than Context-Free (Regular) Kolam Array Grammars. Two subclasses of these grammars are also considered. Comparisons are made. Hierarchies and closure properties are examined. The effects of control devices on GCF(R)KAGs are discussed.

Generative Grammars Without Transformation Rules

Abstract: A phrase-structure grammar has been written which generates exactly the set of sentences generated by a fairly large transformational grammar written by Noam Chomsky.2 The phrase structure version of Chomsky’s grammar is included in the appendis. It is written in an abbreviated notation which is explained below.

Efficient algorithms for automatic construction and compactification of parsing

Abstract: Several computational problems about grammars are studied. Efficient algorithms are presented for the problems of (1) determining, for a given semantic grammar, if there exists a related parsing grammar in some specified grammar class, and (2) finding such a related parsing grammar when one exists. The two grammars are to be related by mergers of nonterminals and/or terminals. Efficient algorithms are presented for most of the grammar classes used in compilers. We also study the problem of (3) determining which terminals of a grammar are good candidates for merger into common lexical tokens of the corresponding parsing grammar.

On the succinctness properties of unordered context-free grammars

Abstract: We prove in this paper that unordered, or ID/LP grammars, are exponentially more succinct than context-free grammars, by exhibiting a sequence (Ln) of finite languages such that the size of any CFG for Ln must grow exponentially in n, but which can be described by polynomial-size ID/LP grammars. The results have implications for the description of free word order languages.

Nonterminal separating Marco grammars

Abstract: We extend the concept of nonterminal separating (or NTS) context-free grammar to nonterminal separating $m$-macro grammar where the mode of derivation $m$ is equal to "unrestricted". "outside-in' or "inside-out". Then we show some (partial) characterization results for these NTS $m$-macro grammars.

Complexity aspects of integrated rewriting—a survey

Abstract: We present an overview of results on the complexity of the membership problem for families of languages generated by several types of generalized grammars. In particular, we consider generalized grammars based on context-independent rewriting, i.e., grammars consisting of a finite number of (non)deterministic substitutions, and on iterated context-dependent rewriting , i.e., grammars composed of a finite number of transductions. We give some conditions on the classes of these substitutions and transductions that guarantee the solvability of this membership problem within certain time and space bounds. As consequences we obtain additional closure properties of some time- and space-bounded complexity classes.

Syntactic approach to image analysis (abstract only)

Abstract: During the past several years, syntactic approach [1,2] has attracted growing attention as promising avenues of approach in image analysis. The object of image analysis is to extract as much information as possible from a given image or a set of images. In this abstract, we will focus our attention on the use of semantic information and grammatical inference.In an attributed grammar, there are still a set of nonterminals, a set of terminals and a start symbol just as in conventional grammars. The productions are different. Each semantic rule. Two kinds of attributes are included in the semantic rules: inherited attributes and synthesized attributes. One example of the attributes is the length of a specific line segment used as a primitive. All the attributes identified for a pattern are expressed in a “total attribute vector”.Instead of using attributes, stochastic grammars associate with each production a probability. That means, one sub-pattern may generate one subpattern with some probability, and another with a different probability. A string may have two or more possible parses. In this case of ambiguity, the probabilities associated with the several possible productions are compared to determine the best fit one. Probabilities are multiplied in multiple steps of stochastic derivations.Besides these, fuzzy languages[3-6] have also been introduced into pattern recognition. By using similarity measures as membership functions, this approach describes patterns in a more understandable way than stochastic grammars. Moreover, fuzzy languages make use of individual characteristics of a class of patterns rather than collective characteristics as in stochastic languages, and therefore it is probably easier to develop grammars than stochastic languages. Yet a lot of work still need to be done in order to develop sufficient theories in this field for practical uses.An appropriate grammar is the core of any type of syntactic pattern recognition process. Grammars may be established by inferring from a priori knowledge about the objects or scenes to be recognized. Another way to establish a pattern grammar is by direct inference from some sample input patterns.Once a grammar is derived from some sample input patterns, other patterns similar to them or belonging to the same class can be parsed according to the grammar. Therefore grammatical inference enables a system to learn most information from an input pattern, and, furthermore, to apply the obtained knowledge to future recognition processes. It seems to be the ultimate aim of image analysis.Inference can be supervised or unsupervised. In supervised inference, a “teacher” who is able to discriminate valid and invalid strings helps in reducing the length of sentences or inserting substrings until some iterative regularity is detected. In unsupervised inference, no prior knowledge about the grammar is assumed.The difficulty of inference is proportional to the complexity of the grammar, and the inference problem does not have a unique solution unless some additional constraints are placed upon the grammars. Some theoretical algorithms have been developed for inferencing regular (finite-state) grammars, but they still have severe limitations for practical use because of large amount computation due to the combinatorial effect.Context-free grammars are even harder to deal with since many decidable properties about regular grammars are undecidable for context-free grammars, such as the equivalency of two contex-free grammars. Therefore, inference algorithms have been developed only for some specific types of context-free grammars and most of them rely on heuristic methods.Syntactic approach to image analysis may be applied to many areas including space object surveillance and identification [7].

SLR(1) and LALR(1) parsing for unrestricted grammars

Abstract: Simple LR(1) and lookahead LR(1) phrase structure grammars are defined and corresponding deterministic two-pushdown automata which parse all sentences are given. These grammars include a wide variety of grammars for non context-free languages. A given phrase structure grammar is one of these types if the parse table for the associated automaton has no multiple entries. A technique for construction of this parse table is given which in the lookahead case involves elimination of inverses in a grammar for lookahead strings for LR(0) items and computation of first sets for strings of symbols in the given grammar.

Graphics and Their Grammars

Abstract: Graphics are graphs with attributes at their vertices. Graphic grammars are natural extensions of graph and attribute grammars with rules that are attributed extensions of the ''pushout'' productions of graph grammars. The theory of graphic grammars is presented and various programming implementations will be discussed. Many motivating examples will be given, including the development of biological organisms the ''semantic net'' representation of expert system knowledge.

Semantics for coordinated substitution grammars as implemented in prolog

Abstract: Syntactic coordination, semantic coordination, and logical meaning generation are all important features of natural language recognition systems. Grammars for natural language, though, rarely describe even syntactic coordination, such as subject-verb number agreement, without the use of additional structures, such as the added arguments in definite clause grammars. The core grammar in such cases simply cannot succinctly represent coordination requirements. A solution is provided by the coordinated substitution grammars (CSubGs), a form of two-level grammar in which indices and their corresponding rules comprise a meta-level grammar used to instantiate rules in the base grammatical form. Through the use of these indices, all syntactic coordination can be concisely and descriptively represented in the grammar. We demonstrate that the CSubGs can generate the recursively enumerable sets, although they can be constrained to generate only the context-free or indexed languages. A program has been developed to translate the grammar rules into Prolog clauses, as is done with definite clause grammars. The remainder of the problem, determining how CSubGs and their indices can affect the natural language semantics in a Prolog implementation, is investigated by studying grammars for various English language constructs. Some aspects of semantic coordination are describable with CSubG indices, although the enforcement of semantic coordination is most generally handled through the use of added nonterminal arguments and additional Prolog clauses. The other facet of natural language semantics, the generation of a logical meaning through the use of added arguments and clauses, is also affected by syntactic coordination index values. Some indices, such as the voice index, cause major restructuring of the logical meaning; others have a smaller effect. Through this work, we demonstrate that coordinated substitution grammars are appropriate for representing natural language syntax and preferable to other formalisms for their robustness and descriptiveness. In addition, some natural language semantics can be represented within the grammar and index values can signal changes in the semantic representation. Finally, we demonstrate a natural language front-end for a Prolog database, indicating the applicability of these results to an actual implementation.

On the writing of grammars

Abstract: Grammar books are written for many purposes and with different users in mind. What are the particular problems of the scholarly reference grammar towards the end of the 20th century?

Multi-level structures for natural language processing

Abstract: This work is a discussion of the advantages of using two-level grammars in syntactic natural language processing. The work contains five main parts. First, a rigorous investigation of the necessary mathematical properties of two-level grammars includes a new type of substitution developed to model government phenomena in natural languages. The second part is a discussion of modeling natural language phenomena in two level grammars. A number of interesting phenomena are considered. We give an improved way of handling gapping and trace phenomena. Crossed serial dependencies, which occur in Swiss German and Dutch, are treated. These are important because they are not context-free. Subcategorization is an integral part of modern linguistic theory and we show how to incorporate this into two-level theory. Thirdly, the work includes a grammar for a large English subset. This grammar models verb phrases with complements, noun phrases with postmodifiers, embedded constructions, conjunctions and prepositional phrases. The ability to handle some ungrammatical sentences is important in computer processing of language. This work shows how two-level grammars give a principled way of allowing the representation of extragrammatical phrases. This is done by relaxing certain constraints which appear in the two-level grammars themselves. This method improves on previous work in syntactic extragrammaticality because it derives the allowed extragrammaticalities directly from the grammar, rather than relying on human intervention. The last part of this dissertation discusses parsing. It shows how to modify context-free algorithms to process two-level grammars. Four implementations are discussed, based on Earley's algorithm and the left corner parsing algorithm. Each algorithm is given in a normal form and in a form which handles extragrammatical input.