Jonathan Dunn

2 paper cards in Mova's library · 2019–2021

Jonathan Dunn wrote 2 of the papers Mova reads, in 2019–2021.

The profile is the library's match of the name; two people with one name are not always told apart.

Papers

Finding Variants for Construction-Based Dialectometry: A Corpus-Based Approach to Regional CxGs (2021)
The paper proposes dialectometry on whole grammars. A construction grammar (4,504 constructions) is first induced from an independent corpus (ukWac), so no variant has to be chosen in advance. Each construction's relative frequency in a sample becomes a feature, and a linear SVM predicts the region of a sample. Two…
Modeling Global Syntactic Variation in English Using Dialect Classification (2019)
The paper asks which English national varieties exist, and then models their syntax. Geo-referenced web text (Common Crawl, 16.65 billion words) and tweets (4.14 billion words) are used; a variety is kept if it has at least 15 million words in both sources, which gives 14 national varieties. The features are…