[Paper Review] Efficient Tabular LR Parsing

Mark-Jan Nederhof, Giorgio Satta|ArXiv.org|May 13, 1996

Natural Language Processing Techniques20 references10 citations

TL;DR

This paper presents a novel, more efficient approach to tabular LR parsing by leveraging grammar transformations and standard chart parsing techniques, significantly reducing both static and dynamic time and space complexity compared to Tomita's generalized LR algorithm. The method achieves superior performance through simplified conceptual foundations and optimized parsing mechanics.

ABSTRACT

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 transformations 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.

Motivation & Objective

• To address the high computational complexity of Tomita's generalized LR parsing algorithm.
• To develop a more conceptually intuitive and efficient parsing framework for context-free grammars.
• To reduce both static and dynamic space and time complexity in tabular LR parsing.
• To demonstrate that grammar transformations and standard tabulation techniques can yield superior performance.
• To provide a simpler, more modular alternative to existing generalized parsing methods.

Proposed method

• The approach uses grammar transformations to simplify the parsing process and eliminate redundant computations.
• It applies standard tabulation techniques (chart parsing) to represent and compute parsing states efficiently.
• The parsing algorithm is designed to avoid the complex data structures and state explosion typical in Tomita's method.
• The method leverages the inherent modularity of grammar transformations to decouple grammar analysis from parsing logic.
• It maintains a compact representation of parsing states using dynamic programming principles.
• The algorithm is formally defined using a set of inference rules that operate on a chart structure.

Experimental results

Research questions

• RQ1 Can grammar transformations and standard tabulation techniques be used to design a more efficient tabular LR parser?
• RQ2 How does the proposed method compare to Tomita's generalized LR algorithm in terms of time and space complexity?
• RQ3 Can a simpler conceptual framework achieve comparable or better performance in generalized parsing?
• RQ4 What is the impact of grammar transformation on the efficiency of parsing ambiguous context-free grammars?
• RQ5 Is it possible to reduce both static and dynamic complexity in tabular LR parsing without sacrificing correctness?

Key findings

• The proposed method achieves significant reductions in both static and dynamic time and space complexity compared to Tomita's algorithm.
• The use of grammar transformations simplifies the parsing process and reduces the number of required parsing states.
• The approach maintains correctness while eliminating the need for complex data structures used in traditional generalized LR parsing.
• The method demonstrates superior efficiency through a more modular and conceptually transparent design.
• The performance improvements are attributed to the combination of grammar transformation and standard chart parsing techniques.
• The algorithm is formally proven to be correct and efficient for arbitrary context-free grammars.