A Corpus-Driven Study of the Variation of Co-Occurrence

A Corpus-Driven Study of the Variation of Co-Occurrence Patterns in Written and Spoken Registers

Abstract
This paper will focus on the study of the variation of co-occurrence patterns
encountered in written and spoken registers, through the analysis of a large lexical
database of corpus-extracted multiword expressions (MWEs) of European Portuguese.
Those MWEs were automatically extracted from a balanced 50 million word written
corpus and a 1 million word spoken corpus, furthermore statistically interpreted using
lexical association measures and partially manually validated in what concerns written
units.
MWEs have been and are still a challenge for linguistic analysis, lexicography
and natural language processing due to their large pattern variation and to the need to
put forward several linguistic levels for their analysis, namely in parameters like degree
of syntactic cohesion, inflected variation and semantic compositional nature.
In this paper, we aim to revise some typologies of MWEs using a corpus-driven
approach, to analyse corpus findings and their relation to MWEs categorization, and to
establish possible contrastive registers based on syntactic, functional and semantic
paradigms: for example, contrasts involving spoken and written texts or contrasts
involving the degree of formality taken transversally in both registers.
By presenting register-specific co-occurrence patterns based on authentic data,
this study will hopefully contribute to the more general categorization of MWEs in
Portuguese.