We present a system for identifying the semantic relationships, or semantic roles, filled by constituents of a sentence within a semantic frame. Various lexical and syntactic features are derived from parse trees and used to derive statistical classifiers from hand-annotated training data. 1

This paper proposes a new theory of the relationship between the sentence processing mechanism and the available computational resources. This theory -- the Syntactic Prediction Locality Theory (SPLT) -- has two components: an integration cost component and a component for the memory cost associated with keeping track of obligatory syntactic requirements. Memory cost is

The Trains project is an effort to build a conversationally proficient planning assistant. A key part of the project is the construction of the Trains system, which provides the research platform for a wide range of issues in natural language understanding, mixedinitiative planning systems, and representing and reasoning about time, actions and events. Four years have now passed since the beginning of the project. Each year we have produced a demonstration system that focused on a dialog that illustrates particular aspects of our research. The commitment to building complete integrated systems is a significant overhead on the research, but we feel it is essential to guarantee that the results constitute real progress in the field. This paper describes the goals of the project, and our experience with the effort so far. This paper is to appear in the Journal of Experimental and Theoretical AI, 1995. The TRAINS project has been funded in part by ONR/ARPA grant N00014-92-J-1512, U.S. Air ...

this paper is to introduce, by means of the detailed analysis of a single grammatical problem, the rudiments of a grammatical theory which assigns a central role to the notion of grammatical construction. To adopt a constructional approach is to undertake a commitment in principle to account for the entirety of each language. 2 This means that the relatively general patterns of the language, such as the one licensing the ordering of a finite auxiliary verb before its subject in English as illustrated in (1), and the more idiomatic patterns, such as those exemplified in (2), stand on an equal footing as data for which the grammar must provide an account. (1) a What have you done? b Never will I leave you. c So will she. d Long may you prosper! e Had I known, . . . f Am I tired! g . . . as were the others h Thus did the hen reward Beecher. (2) a by and large b [to] have a field day c [to] have to hand it to [someone] d (*A/*The) Fool that I was, . . . e in x's own right Given such a commitment, the construction grammarian is required to develop an explicit system of representation, capable of encoding economically and without loss of generalization, all the constructions (or patterns) of the language, from the most idiomatic to the most general. This goal was advanced in the form of a promissory note in an earlier paper that dealt with the English let alone construction: "It appears to us that the machinery needed for describing the so-called minor or peripheral constructions of the sort which has occupied us here will have 1 The authors gratefully acknowledge much fruitful discussion regarding the content of this paper with Mary Catherine O'Connor. We are indebted to Yunsook Chung, Ron Kaplan, Ray Jackendoff, Susanne Riehemann and Ivan Sag for comments on earlier dr...

We present a critical overview of the state-of-the-art in parser evaluation methodologies and metrics. A discussion of their relative strengths and weaknesses motivates a new---and we claim more informative and generally applicable---technique of measuring parser accuracy, based on the use of grammatical relations. We conclude with some preliminary results of experiments in which we use this new scheme to evaluate a robust parser of English.

This article shows that the Web can be employed to obtain frequencies for bigrams that are unseen in a given corpus. We describe a method for retrieving counts for adjective-noun, noun-noun, and verb-object bigrams from the Web by querying a search engine. We evaluate this method by demonstrating: (a) a high correlation between Web frequencies and corpus frequencies; (b) a reliable correlation between Web frequencies and plausibility judgments; (c) a reliable correlation between Web frequencies and frequencies recreated using class-based smoothing; (d) a good performance of Web frequencies in a pseudodisambiguation task. 1.

The problems of access -- retrieving linguistic structure from some mental grammar -- and disambiguation -- choosing among these structures to correctly parse ambiguous linguistic input -- are fundamental to language understanding. The literature abounds with psychological results on lexical access, the access of idioms, syntactic rule access, parsing preferences, syntactic disambiguation, and the processing of garden-path sentences. Unfortunately, it has been difficult to combine models which account for these results to build a general, uniform model of access and disambiguation at the lexical, idiomatic, and syntactic levels. For example psycholinguistic theories of lexical access and idiom access and parsing theories of syntactic rule access have almost no commonality in methodology or coverage of psycholinguistic data. This paper presents a single probabilistic algorithm which models both the access and disambiguation of linguistic knowledge. The algorithm is based on a parallel parser which ranks constructions for access, and interpretations for disambiguation, by their conditional probability. Low-ranked constructions and interpretations are pruned through beam-search; this pruning accounts, among other things, for the garden-path effect. I show that this motivated probabilistic treatment accounts for a wide variety of psycholinguistic results, arguing for a more uniform representation of linguistic knowledge and for the use of probabilisticallyenriched grammars and interpreters as models of human knowledge of and processing of language.

The unification problem and several variants are presented. Various algorithms and data structures are discussed. Research on unification arising in several areas of computer science is surveyed, these areas include theorem proving, logic programming, and natural language processing. Sections of the paper include examples that highlight particular uses

. Though [Pollard and Sag 1994] assumes that an unspecified variant of the formal logic of [Carpenter 1992] will provide a formalism for HPSG, a precise formulation of the envisaged formalism is not immediately obvious, primarily because a principal tenet of [Carpenter 1992], that feature structures represent partial information, seems to conflict with a principal tenet of [Pollard and Sag 1994], that feature structures represent abstract linguistic entities. This has caused many HPSGians to be mistakenly concerned with partial-information specific notions, such as subsumption, that are appropriate for the [Carpenter 1992] logic but inappropriate for the formalism [Pollard and Sag 1994] envisages. This paper hopes to allay this concern and the confusion it engenders by substituting [King 1989] for [Carpenter 1992] as the basis of the envisaged formalism. It demonstrates that the formal logic of [King 1989] provides a formalism for HPSG that meets all [Pollard and Sag 1994] asks of the ...

Abstract not found