2004 · 13 citations · 4 references
Language StudiesComputer ScienceComputational LinguisticsNatural Language ProcessingSemanticsAnnotationText Similarity AnalysisDisambiguation Evaluation MetricsSemantic DisambiguationSemantic SimilaritySense DisambiguationSimilarity Measure Selection
This paper presents an unsupervised method for the resolution of lexical ambiguity of nouns. The method relies on the topological structure of the noun taxonomy of WordNet where a notion of semantic distance is defined. An unsupervised semantic tagger, based on the above measure, is evaluated over an hand-annotated portion of the British National Corpus and compared with a supervised approach based on the Maximum Entropy Model. 1.
4
Unsupervised word sense disambiguation rivaling supervised methods
David Yarowsky · 1995 · 2.4K citations · Full text
Hiding a Semantic Hierarchy in a Markov Model
Steven Abney, Marc Light · 1999 · 43 citations
Computing, Computer Science, Natural Language Processing +13