Unification grammars.

From MaRDI portal





This book provides a theoretically well-founded introduction to (untyped) unification grammars, one of the most influential grammar theory approaches within the field of computational linguistics. The book focuses on two dimensions of grammar description, the formal foudations underlying unification grammars and their computational implications as a basis for reasoning about grammars, on the one hand, and fundamental natural language syntactic constructions and the way they are specified in unification grammars, on the other hand. Single chapters of the book deal with fundamental concepts underlying unification grammars (such as context-free grammars (CFGs) and the relationship between grammars and languages) and the main building blocks of unification grammars, such as feature structures, feature graphs, feature graph subsumption, attribute-value matrices and the unification operation for feature structures and feature graphs, which combines the information that is contained in two (compatible) feature structures. Various grammar fragments are presented for the syntactic description of natural language phenomena (for English) such as number agreement, case control, subcategorization, long-distance dependencies, relative clauses, subject/object control, constituent coordination. Finally, computational issues, i.e., parsability of and parsing with unification grammars, are dealt with as well, including considerations of the greater generative capacity (formal expressiveness) of unification grammars compared with CFGs and the tractability and complexity of parsing algorithms for unification grammars. An appendix contains solutions of selected exercises presented in the various chapters of the book.











This page was built for publication: Unification grammars.

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2891521)