Ide and Veronis provide a good overview of the issue (termed "
word sense disambiguation" in computational linguistics, from 1998 (Ide, 1998).
Among the topics are
word sense disambiguation, domain adaptation in part-of-speech tagging, a computational cognitive model of human translation processes, interactive question answering, mining user-generated content for social research and other applications, and spreadsheeting using modified visual feature vectors.
In their introduction to a recent monograph on WSD, Agirre and Edmonds suggested that one of the ways to achieve fundamentally new research results in the future is to "become more theoretical [...], to work on WSD embracing more realistic models of
word sense (including non-discreteness, vagueness, and analogy), thus drawing on and feeding theories of word meaning and context from (computational) lexical semantics and lexicography" (Agirre & Edmonds, 2006: 17).
The Asian contributors propose
word sense disambiguation based on relation structure, a remote sensing image fusion algorithm, strategies of modeling from VDM-SL to JML, and a negotiation management method for the IKE protocol.
The fun I found in this game was in using my
word sense in a different way.
White in [11] describes a
word sense disambiguation scheme that depends on a sort of "minimalist" rendering of the dictionary: The senses in the MRD are associated only with their taxonomic structure, as derived from the tangled hierarchy procedures mentioned, and collocationally with the words of their own definitions.
In the analysis phase a series of modules (tokenizer, morphological analyzer, part of speech tagger, chunker, named entity recognizer, parser, and
word sense disambiguator) generate an intermediate representation that is easy to transfer (syntax transfer, lexical transfer, and some transliteration).
(3) The task of distinguishing between the meanings of a polysemous word, is called
word sense disambiguation(WSD).
As regards language teaching, the context-dependency of
word senses raises the question of how useful it is to learn a given
word sense without its corresponding co-textual correlates.