Title: The representation of polysemy through vectors: some building blocks for constructing models and applications with LSA

Authors: Guillermo Jorge-Botana, Jose A. Leon, Ricardo Olmos, Inmaculada Escudero

Addresses: Departamento de Psicologia Basica, Facultad de Psicologia, Universidad Autonoma de Madrid, Campus de Cantoblanco, 28049 – Madrid, Spain. ' Departamento de Psicologia Basica, Facultad de Psicologia, Universidad Autonoma de Madrid, Campus de Cantoblanco, 28049 – Madrid, Spain. ' Departamento de Psicologia Basica, Facultad de Psicologia, Universidad Autonoma de Madrid, Campus de Cantoblanco, 28049 – Madrid, Spain. ' Departamento de Psicologia Evolutiva y de la Educacion, Facultad de Psicologia, UNED, C/ Juan del Rosal, 10, 28040 – Madrid, Spain

Abstract: The problem of the multiplicity of word meanings has preoccupied so many researches from the linguistics, psychology or computational linguistic. In this paper, we revised how LSA represents the polysemous words and we explain some bias related with the meaning generation and revised some constraint-satisfaction models which introduce into the equation some dynamic mechanisms. The idea of these models is to take the amalgamated word vector from LSA and embed it into its discourse and semantic context, and by means of a dynamic mechanism, the appropriate features of it is are selected. To illustrate our arguments, we present some networks, providing evidence that polysemous words have separated representations for each sense only in presence of the linguistic context that involved it. We also present an example of how these mechanisms also contribute to support the visual heuristic searches in the visual information retrieval interfaces (VIRIs).

Keywords: latent semantic analysis; LSA; polysemy; lexical representation; linguistic contexts; lexical semantics; computational linguistics; predication; discourse comprehension; psycholinguistics; ambiguity; semantic networks; visual information retrieval interfaces; VIRIs; word meanings; polysemous words; constraint-satisfaction models; dynamic mechanisms; amalgamated vectors; word vectors; visual searches; heuristic searches; word sense; visualisation; continuing education; life-long learning; automatic free-text evaluation.

DOI: 10.1504/IJCEELL.2011.042791

International Journal of Continuing Engineering Education and Life-Long Learning, 2011 Vol.21 No.4, pp.328 - 342

Published online: 04 Oct 2011 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article