ANÁLISIS DE LAS RELACIONES SEMÁNTICAS A TRAVÉS DE UNA TAREA DE LIBRE ASOCIACIÓN EN ESPAÑOL CON MAPAS AUTO-ORGANIZADOS

Abstract

In Psycholinguistics and Computational Linguistics free association tasks have been used commonly as general data collection to develop semantic networks that simulate semantic space of the speakers of a language. Semantic networks incorporate measures such as number of associates, associative strength and other groups of quantitative features of diverse entity. Nevertheless, the number of studies that considers a word from the point of view of the sort of association that it evokes is less abundant. This paper analyzes semantic-associative relationships of 150 Spanish words obtained from a discrete free-association task with the aim of proposing a word taxonomy based on the self-organizing maps (SOM). This procedure will generate both a set of semantic variables based on associates and word taxonomy to be used as independent variables in cognitive psychology experiments that focus on how associative semantic features of words predict linguistic performance.