Cognitive Linguistic forensic authorship analysis using the likelihood ratio framework

Nini (2023a) proposes a Theory of Linguistic Individuality based on Cognitive Linguistics and Cognitive Psychology. The central argument of this theory is that each individual possesses a unique repertoire of linguistic units, defined following Langacker (1987) as structures that a person can produce automatically and that are stored as traces of procedural memory. Nini (2023a) then proposes methods compatible with this theory that outperform traditional computational methods based on frequency of features. These novel set-theory methods are generalisations of n-gram tracing (Grieve et al. 2019).
In this talk, the application of these techniques is demonstrated with tests on a set of ten corpora that simulate various forensic authorship verification scenarios, from emails to academic papers, including cross-domain problems. The results of these analyses show that the method proposed in Nini (2023) outperforms state of the art methods in authorship verification while also being more explorable by a human analyst and compatible with the likelihood ratio framework. The talk also introduces the R package “idiolect” to conduct these analyses (Nini 2023b).

References
Grieve, Jack, Emily Chiang, Isobelle Clarke, Hannah Gideon, Aninna Heini, Andrea Nini & Emily Waibel. 2019. Attributing the Bixby Letter using n-gram tracing. Digital Scholarship in the Humanities 34(3). 493–512.
Langacker, Ronald W. 1987. Foundations of cognitive grammar. Vol. 1. Stanford, CA: Stanford University Press.
Nini, Andrea. 2023a. A Theory of Linguistic Individuality for Authorship Analysis (Elements in Forensic Linguistics). Cambridge, UK: Cambridge University Press.
Nini, Andrea. 2023b. Idiolect: An R package for forensic authorship analysis. https://github.com/andreanini/idiolect.