Open Set Authorship Attribution toward Demystifying Victorian Periodicals

TL;DR

This paper explores authorship attribution in Victorian literature, highlighting the challenges of open-set classification with many candidate authors and evaluating the effectiveness of standard machine learning methods.

Contribution

It introduces a new dataset for Victorian texts and analyzes the limitations of existing methods in open-set authorship attribution scenarios.

Findings

Linear classifiers perform well in closed-set attribution.

Standard methods struggle with large candidate pools in open-set scenarios.

Robust approaches are needed for real-world authorship attribution.

Abstract

Existing research in computational authorship attribution (AA) has primarily focused on attribution tasks with a limited number of authors in a closed-set configuration. This restricted set-up is far from being realistic in dealing with highly entangled real-world AA tasks that involve a large number of candidate authors for attribution during test time. In this paper, we study AA in historical texts using anew data set compiled from the Victorian literature. We investigate the predictive capacity of most common English words in distinguishing writings of most prominent Victorian novelists. We challenged the closed-set classification assumption and discussed the limitations of standard machine learning techniques in dealing with the open set AA task. Our experiments suggest that a linear classifier can achieve near perfect attribution accuracy under closed set assumption yet, the need for more robust approaches becomes evident once a large candidate pool has to be considered in the open-set classification setting.