Visualizing distances in a set of near-synonyms [chapter]

Guillaume Desagulier
2014 Human Cognitive Processing  
I aim to uncover revealing aspects of the conceptual structure of four English moderators (rather, quite, fairly, and pretty) and shed new light on previous studies on degree modifiers. I develop an original methodology for handling and visualizing measures of significant attraction between lexical items. This methodology combines univariate and multivariate statistics. Collexeme analysis is used as input for hierarchical agglomerative cluster analysis, and multiple distinctive collexeme
more » ... s is used as input for correspondence analysis. Visualizing collexemes with exploratory tools does more than depict proximities and distances between individuals and variables. It is also an accurate means to (a) unveil fine semantic differences in a set of near-synonymous constructions, (b) determine entrenchment continua, and (c) represent a significant part of the complex inventory of intensifying constructions.
doi:10.1075/hcp.43.06des fatcat:zdchv7fvfzhdncbwk7gmkiyh5a