We develop an improved form of left-corner chart parsing for large context-free grammars, introducing improvements that result in significant speed-ups compared to previously-known variants of left-corner parsing. We also compare our method to several other major parsing approaches, and find that our improved left-corner parsing method outperforms each of these across a range of grammars. Finally, we also describe a new technique for minimizing the extra information needed to efficiently recover parses from the data structures built in the course of parsing. 1

In this paper we relate a number of parsing algorithms which have been developed in very different areas of parsing theory, and which include deterministic algorithms, tabular algorithms, and a parallel algorithm. We show that these algorithms are based on the same underlying ideas.

We give a new treatment of tabular LR parsing, which is an alternative to Tomita's generalized LR algorithm. The advantage is twofold. Firstly, our treatment is conceptually more attractive because it uses simpler concepts, such as grammar trans- formations and standard tabulation techniques also know as chart parsing. Secondly, the static and dynamic complexity of parsing, both in space and time, is significantly reduced.

Abstract not found

We present new results on the relation between purely symbolic contextfree parsing strategies and their probabilistic counter-parts. Such parsing strategies are seen as constructions of push-down devices from grammars. We show that preservation of probability distribution is possible under two conditions, viz. the correct-prefix property and the property of strong predictiveness. These results generalize existing results in the literature that were obtained by considering parsing strategies in isolation. From our general results we also derive negative results on so-called generalized LR parsing. 1

This report describes research done in the context of a subproject of the HPCN project IMPACT. The IMPACT project is headed by the ING bank and is founded by the organization for High Performance Computing and Networking (HPCN). The aim of the specific subproject, in the context of which this report has been written, is to develop (techniques for) natural language interfaces to information resources, focusing on the use of high-performance computers to achieve acceptable response times. This report is part of the "Parallel Parsing I" research topic. IMPACT-NLI-1997-1 ii Preface IMPACT IMPACT-NLI-1997-1 IMPACT iii Contents Preface i 1 Introduction 1 2 Basics 3

In this paper we describe the ideas behind the Grammar Workbench (GWB). The GWB is one of a series of tools for the development of AGFLs (affix grammars over a finite lattice) for natural languages. Its functions comprise a specialised editor, computation of properties, a random generator of sentences, and special functions to provide an overview of a grammar. This paper discusses the functions of the GWB, the grammatical formalism AGFLs, and the AGFL project. We also discuss the relationship between the complete development environment for AGFLs and other development environments, both for other grammatical formalisms and for computer programs. 1 Introduction Formal grammars for natural languages tend to become unmanageable as they get larger. Similar problems occur in the development of large computer programs. To overcome these problems with large programs, a number of techniques have been invented in the field of software engineering to ffl structure programs so as to permit deco...

This paper deals with affix grammars over a finite lattice (AGFLs). The parameters in AGFLs are called affixes. AGFLs are a simple formalism but have still been proved powerful enough for the description of various natural languages [20]. Context-free parsing can be implemented very efficiently by constructing parse forests. A parse forest is a concise representation of a number of parse trees for an ambiguous sentence. Parse forests can be constructed in polynomial time and require polynomial space for storage, measured in the length of the input sentence. In this paper, the extra parameters of extended context-free grammars are used to impose restrictions (called context dependencies) on the allowable parse trees. Context-free parsing must therefore be augmented to calculate the parameters. Affix values for AGFLs can be evaluated efficiently because the domains are finite. The rejection of parse trees based on violated context dependencies is more complicated if parse trees are merged into parse forests. Because of the ambiguous nature of natural language, grammars describing natural languages are necessarily also ambiguous, and therefore context dependencies cannot determine a unique parse tree for every sentence. Instead, some humancomputer interaction is necessary to find the intended parses of sentences.

Head-Corner (HC) parsing has come up in computational linguistics a few years ago, motivated by linguistic arguments. This idea is a heuristic, rather than a fail-safe principle, hence it is relevant indeed to consider the worst-case behaviour of the HC parser. We define a novel predictive head-corner chart parser of cubic time complexity. We start with a left-corner (LC) chart parser, which is easier to understand. Subsequently, the LC chart parser is generalized to an HC chart parser. It is briefly sketched how the parser can be enhanced with feature structures. 1. Introduction "Our Latin teachers were apparently right", Martin Kay (1989) remarks. "You should start [parsing] with the main verb. This will tell you what kinds of subjects and objects to look for and what cases they will be in. When you come to look for these, you should also start by trying to find the main word, because this will tell you most about what else to look for". Head-driven or head-corner parsing has been ...

In this paper, we present a method which, in practice, allows to use parsers for languages defined by very large context-free grammars (over a million symbol occurrences). The idea is to split the parsing process in two passes. A first pass computes a sub-grammar which is a specialized part of the large grammar selected by the input text and various filtering strategies. The second pass is a traditional parser which works with the subgrammar and the input text. This approach is validated by practical experiments performed on a Earley-like parser running on a test set with two large context-free grammars. 1