Probabilistic Frame-Semantic Parsing

Human Language Technologies: The 2010 Annual Conference of the North American Chapter of the ACL, pages 948–956, Los Angeles, California, June 2010. c⃝2010 Association for Computational Linguistics Probabilistic Frame-Semantic Parsing Dipanjan Das Nathan Schneider Desai Chen Noah A. Smith School of Computer Science Carnegie Mellon University Pittsburgh, PA 15213, USA {dipanjan@cs,nschneid@cs,desaic@andrew,nasmith@cs}.cmu.edu Abstract This paper contributes a formalization of frame-semantic parsing as a structure predic- tion problem and describes an implemented parser that transforms an English sentence into a frame-semantic representation. It finds words that evoke FrameNet frames, selects frames for them, and locates the arguments for each frame. The system uses two feature- based, discriminative probabilistic (log-linear) models, one with latent variables to permit disambiguation of new predicate words. The parser is demonstrated to significantly outper- form previously published results. 1 Introduction FrameNet (Fillmore et al., 2003) is a rich linguistic resource containing considerable information about lexical and predicate-argument semantics in En- glish. Grounded in the theory of frame semantics (Fillmore, 1982), it suggests—but does not formally define—a semantic representation that blends word- sense disambiguation and semantic role labeling. In this paper, we present a computational and statistical model for frame-semantic parsing, the problem of extracting from text semantic predicate- argument structures such as those shown in Fig. 1. We aim to predict a frame-semantic representation as a structure, not as a pipeline of classifiers. We use a probabilistic framework that cleanly integrates the FrameNet lexicon and (currently very limited) available training data. Although our models often involve strong independence assumptions, the prob- abilistic framework we adopt is highly amenable to future extension through new features, relaxed in- dependence assumptions, and semisupervised learn- ing. Some novel aspects of our current approach include a latent-variable model that permits disam- biguation of words not in the FrameNet lexicon, a unified model for finding and labeling arguments, TRANSITIVE_ ACTION Agent Patient Event Cause Place Time CAUSE_TO_ MAKE_NOISE Agent Sound_maker Cause Place Time MAKE_NOISE Noisy_event Sound Sound_source Place Time cough.v, gobble.v, ring.v, yodel.v, ... blare.v, play.v, ring.v, toot.v, ... — Inheritance relation Causative_of relation Excludes relation Purpose Figure 2. Partial illustration of frames, roles, and LUs related to the CAUSE TO MAKE NOISE frame, from the FrameNet lexicon. “Core” roles are filled ovals. 8 addi- tional roles of CAUSE TO MAKE NOISE are not shown. and a precision-boosting constraint that forbids ar- guments of the same predicate to overlap. Our parser achieves the best published results to date on the SemEval’07 FrameNet task (Baker et al., 2007). 2 Resources and Task We consider frame-semantic parsing resources. 2.1 FrameNet Lexicon The FrameNet lexicon is a taxonomy of manu- ally identified general-purpose frames for English.1 Listed in the lexicon with each frame are several lemmas (with part of speech) that can denote the frame or some aspect of it—these are called lexi- cal units (LUs). In a sentence, word or phrase to- kens that evoke a frame are known as targets. The set of LUs listed for a frame in FrameNet may not be exhaustive; we …

Content truncated to 500 words