Dale Gerdemann

Finite State Transducers with Predicates and Identities

An extension to finite state transducers is presented, in which atomic symbols are replaced by arbitrary predicates over symbols. The extension is motivated by applications in natural language processing (but may be more widely applicable) as well as by the observation that transducers with predicates generally have fewer states and fewer transitions. Although the extension is fairly trivial for finite state acceptors, the introduction of predicates is more interesting for transducers. It is shown how various operations on transducers (e.g., composition) can be implemented, as well as how the transducer determinization algorithm can be generalized for predicate-augmented finite state transducers.

Approximation and Exactness in Finite State Optimality Theory

Previous work (Frank and Satta 1998; Karttunen, 1998) has shown that Optimality Theory with gradient constraints generally is not finite state. A new finite-state treatment of gradient constraints is presented which improves upon the approximation of Karttunen (1998). The method turns out to be exact, and very compact, for the syllabification analysis of Prince and Smolensky (1993).

Transducers from rewrite rules with backreferences

Context sensitive rewrite rules have been widely used in several areas of natural language processing, including syntax, morphology, phonology and speech processing. Kaplan and Kay, Karttunen, and Mohri & Sproat have given various algorithms to compile such rewrite rules into finite-state transducers. The present paper extends this work by allowing a limited form of backreferencing in such rules. The explicit use of backreferencing leads to more elegant and general solutions.

The correct and efficient implementation of appropriateness specifications for typed feature structures

In this paper, we argue that type inferencing incorrectly implements appropriateness specifications for typed feature structures, promote a combination of type resolution and unfilling as a correct and efficient alternative, and consider the expressive limits of this alternative approach. Throughout, we use feature cooccurence restrictions as illustration and linguistic motivation.

Direct automated inversion of logic grammars

Reversibility of logic grammars in natural language processing is desirable for both theoretical and practical reasons. This paper addresses this topic in describing a new approach to automated inversion of logic grammars: the Direct Inversion Approach (dia). A logic grammar is inverted by automatically altering the order of literals in the grammar and reformulating certain recursive procedures at compile time. The inversion process results in a new executable grammar, which is evaluated top-down and left-to-right (using a standard Prolog interpreter), but not left-to-right with respect to the original grammar. Thedia improves upon related approaches not only in being fully automated and computationally tractable, but also with respect to the class of grammars it is able to invert and the performance of the new executable grammar produced.

Parsing as tree traversal

This paper presents a unified approach to parsing, in which top-down, bottom-up and left-corner parsers are related to preorder, postorder and inorder tree traversals. It is shown that the simplest bottom-up and left-corner parsers are left recursive and must be converted using an extended Greibach normal form. With further partial execution, the bottom-up and left-corner parsers collapse together as in the BUP parser of Matsumoto.

Off-line optimization for Earley-style HPSG processing

A novel approach to HPSG based natural language processing is described that uses an off-line compiler to automatically prime a declarative grammar for generation or parsing, and inputs the primed grammar to an advanced Earley-style processor. This way we provide an elegant solution to the problems with empty heads and efficient bidirectional processing which is illustrated for the special case of HPSG generation. Extensive testing with a large HPSG grammar revealed some important constraints on the form of the grammar.