An Image Analysis Approach to the Calligraphy of Books

TL;DR

This paper explores how image analysis of visualized text networks can be used for authorship attribution, showing that combining geometrical and topological features enhances accuracy.

Contribution

It introduces a method that integrates image analysis with topological network features for authorship attribution, demonstrating improved performance over previous approaches.

Findings

Visual features perform similarly to topological features in authorship attribution.

Combining visual and topological features improves attribution accuracy.

Image analysis techniques effectively quantify geometrical properties of text networks.

Abstract

Text network analysis has received increasing attention as a consequence of its wide range of applications. In this work, we extend a previous work founded on the study of topological features of mesoscopic networks. Here, the geometrical properties of visualized networks are quantified in terms of several image analysis techniques and used as subsidies for authorship attribution. It was found that the visual features account for performance similar to that achieved by using topological measurements. In addition, the combination of these two types of features improved the performance.