Search

International audience; Multifactorial usage-feature analysis (profile-based approach) has been successfully applied to polysemy research (Gries 2006; Glynn 2009, 2010). This chapter represents a repeat analysis of Gries (2006). The study has three aims: (i) to verify the results of the previous study; (ii) to identify limitations in the applications of the statistical technique employed (hierarchical cluster analysis) in the previous study; and (iii) to demonstrate the need to account for sociolinguistic dimensions in polysemy research. The study is based on a sample of 500 occurrences of the lexeme to run, extracted in even proportions from British English and American English and from online personal journals (blogs) and conversations (American National Corpus and British National Corpus).

International audience

International audience; Correspondence analysis is an exploratory technique for complex categorical data, typical of corpus-driven research. It identifies patterns of association and disassociation in those data. For instance, it can map the correlations between different uses of a linguistic form and its various social and/or morpho-syntactic contexts. The technique presents its results in the form of a two-dimensional plot, which visualises these relationships in an intuitive manner. These plots offer rich representations of the relations between different facets of complex data. Using R, this chapter explains how the technique works and offers a step-by-step explanation of its application and the interpretation of its results. The technique is also compared to the better-known and comparable cluster analysis.

International audience; The use of corpora in semantic research is a rapidly developing method. However, the range of quantitative techniques employed in the field can make it difficult for the non-specialist to keep abreast with the methodological development. This chapter serves as an introduction to the use of corpus methods in Cognitive Semantic research and as an overview of the relevant statistical techniques and software needed for performing them. The discussion and description are intended for researches in semantics that are interested in adopting quantitative corpus-driven methods. The discussion argues that there are fundamentally two corpus-driven approaches to meaning, one based on observable formal patterns (collocation analysis) and another based on patterns of annotated usage-features of use (feature analysis). The discussion then introduces and explains each of the statistical techniques currently used in the field. Examples of the use of each technique are listed and a summary of the software packages available in R for performing the techniques is included.

International audience; This chapter introduces the field of polysemy and synonymy studies from a Cognitive Linguistic perspective. Firstly, the discussion explains and defines the object of research, showing that the study of semantic relations, traditionally restricted to the description of lexical semantics, needs to be extended to include all formal structures, including morpho-syntax. Secondly, given the theoretical assumptions of Cognitive Linguistics, it is argued that quantitative corpus-driven methods are essential for the description of semantic structures. Lastly, the chapter charts the development of Cognitive Semantic research in polysemy and synonymy and demonstrates how the current corpus-driven research in the field is inherently linked to the traditions of radial network analysis and prototype semantics. It is argued that instead of an empirical revolution (as has been suggested in recent commentaries), the current trends in the use of observational data are a natural extension of the Cognitive Semantic research tradition