Enhancing Representation Generalization in Authorship Identification

TL;DR

This paper reviews methods for improving the generalization of stylistic features in authorship identification, emphasizing the importance of feature selection and deep learning approaches for cross-domain accuracy.

Contribution

It systematically analyzes stylistic features and strategies to enhance cross-domain generalization in authorship identification, highlighting the effectiveness of deep learning models.

Findings

Character n-grams and function words are robust features.

Content words can bias and reduce cross-domain accuracy.

Deep learning models with syntactic info improve generalization.

Abstract

Authorship identification ascertains the authorship of texts whose origins remain undisclosed. That authorship identification techniques work as reliably as they do has been attributed to the fact that authorial style is properly captured and represented. Although modern authorship identification methods have evolved significantly over the years and have proven effective in distinguishing authorial styles, the generalization of stylistic features across domains has not been systematically reviewed. The presented work addresses the challenge of enhancing the generalization of stylistic representations in authorship identification, particularly when there are discrepancies between training and testing samples. A comprehensive review of empirical studies was conducted, focusing on various stylistic features and their effectiveness in representing an author's style. The influencing factors such as topic, genre, and register on writing style were also explored, along with strategies to mitigate their impact. While some stylistic features, like character n-grams and function words, have proven to be robust and discriminative, others, such as content words, can introduce biases and hinder cross-domain generalization. Representations learned using deep learning models, especially those incorporating character n-grams and syntactic information, show promise in enhancing representation generalization. The findings underscore the importance of selecting appropriate stylistic features for authorship identification, especially in cross-domain scenarios. The recognition of the strengths and weaknesses of various linguistic features paves the way for more accurate authorship identification in diverse contexts.