A Framework for Authorial Clustering of Shorter Texts in Latent Semantic Spaces

Abstract

Authorial clustering involves the grouping of documents written by the same author or team of authors without any prior positive examples of an author's writing style or thematic preferences. For authorial clustering on shorter texts (paragraph-length texts that are typically shorter than conventional documents), the document representation is particularly important. We propose a high-level framework which utilizes a compact data representation in a latent feature space derived with nonparametric topic modeling. Authorial clusters are identified thereafter in two scenarios: (a) fully unsupervised and (b) semi-supervised where a small number of shorter texts are known to belong to the same author (must-link constraints) or not (cannot-link constraints).We report on experiments with 120 collections in three languages and two genres and show that the topic-based latent feature space provides a promising level of performance while reducing the dimensionality by a factor of 1500 compared to state-of-the-art. We also demonstrate that little knowledge on constraints in authorial clusters memberships leads to auspicious improvements in front of this difficult task.

LAC: Latent Authorial Clustering of shorter texts

“…To the convenience of researchers, LAC can seamlessly work on a set of corpora, and it integrates the results of all datasets in one final output to enable more reproducible and reliable reporting of performance levels. Till now, LAC has enabled the work done by Trad and Spiliopoulou [10]. A next step would be to devise a customised similarity metric by which we can assess how similar two documents are in terms of writing style.…”