Year: 2009
Type: conference-paper
Abstract: This research examines a word sense disambiguation method using selectors acquired from the Web. Selectors describe words which may take the place of another given word within its local context. Work in using Web selectors for noun sense disambiguation is generalized into the disambiguation of verbs, adverbs, and adjectives as well. Additionally, this work incorporates previously ignored adverb context selectors and explores the effectiveness of each type of context selector according to its part of speech. Overall results for verb, adjective, and adverb disambiguation are well above a random baseline and slightly below the most frequent sense baseline, a point which noun sense disambiguation overcomes. Our experiments find that, for noun and verb sense disambiguation tasks, each type of context selector may assist target selectors in disambiguation. Finally, these experiments also help to draw insights about the future direction of similar research.
Authors: Hansen A. Schwartz, Fernando Gomez
Institution: University of Central Florida
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
OpenAlex ID: W1976474174