Register analysis in blogs: Correlation between professional sector and functional dimensions
Year: 2013
Type: report
Abstract: This study uses multi-dimensional analysis to investigate the distribution of a varied set of linguistic features in blogs, with the goal of finding dimensions of functional variation that allow to distinguish between two opposing occupational sectors, the Humanities and Exact Sciences. A corpus of 9 million words sampled across a wide range of topics and written by men in their twenties is tagged and functions as input for a frequency matrix, on which a factor analysis is carried out to identify quantitatively the co-occurrence relations among linguistic features. The interpretation of the four resultant factors reveals clear axes of variation, allowing to some extent to differentiate the professional sectors. Furthermore, insight is gained into certain functional tendencies across dimension borders.
Source: Lirias (KU Leuven)
Authors: Jocelyne Daems, Dirk Speelman, Tom Ruette
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2519269330