Synchronous Context-Free Grammars and Optimal Linear Parsing Strategies
Pierluigi Crescenzi, Daniel Gildea, Andrea Marino, Gianluca Del Rossi, Giorgio Satta
Abstract
Pierluigi Crescenzi, Daniel Gildea, Andrea Marino, Gianluca Del Rossi, Giorgio Satta
Abstract
Synchronous Context-Free Grammars (SCFGs), also known as syntax-directed translation schemata, are unlike context-free grammars in that they do not have a binary normal form. In general, parsing with SCFGs takes space and time polynomial in the length of the input strings, but with the degree of the polynomial depending on the permutations of the SCFG rules. We consider linear parsing strategies, which add one nonterminal at a time. We show that for a given input permutation, the problems of finding the linear parsing strategy with the minimum space and time complexity are both NP-hard.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Synchronous Context-Free Grammars (SCFGs), also known as syntax-directed translation schemata, are unlike context-free grammars in that they do not have a binary normal form. In general, parsing with SCFGs takes space and time polynomial in the length of the input strings, but with the degree of the polynomial depending on the permutations of the SCFG rules. We consider linear parsing strategies, which add one nonterminal at a time. We show that for a given input permutation, the problems of finding the linear parsing strategy with the minimum space and time complexity are both NP-hard.
Key concepts: Terminal and nonterminal symbols, Context-free grammar, Indexed grammar, Parsing expression grammar, Parsing, Synchronous context-free grammar, L-attributed grammar, Computer science