Text authorship identified using the dynamics of word co-occurrence networks

Abstract

The identification of authorship in disputed documents still requires human expertise, which is now unfeasible for many tasks owing to the large volumes of text and authors in practical applications. In this study, we introduce a methodology based on the dynamics of word co-occurrence networks representing written texts to classify a corpus of 80 texts by 8 authors. The texts were divided into sections with equal number of linguistic tokens, from which time series were created for 12 topological metrics. The series were proven to be stationary (p-value>0.05), which permits to use distribution moments as learning attributes. With an optimized supervised learning procedure using a Radial Basis Function Network, 68 out of 80 texts were correctly classified, i.e. a remarkable 85% author matching success rate. Therefore, fluctuations in purely dynamic network metrics were found to characterize authorship, thus opening the way for the description of texts in terms of small evolving networks. Moreover, the approach introduced allows for comparison of texts with diverse characteristics in a simple, fast fashion.

References in corpus (23)

• Emergence of scaling in random networks
• Why social networks are different from other types of networks
• Global Organization of the Lexicon
• A systematic comparison of supervised classifiers
• Limited Urban Growth: London's Street Network Dynamics since the 18th Century
• Criticality of spreading dynamics in hierarchical cluster networks without inhibition
• A complex network approach to stylometry
• Probing the topological properties of complex networks modeling short written texts
• Probing the statistical properties of unknown texts: application to the Voynich Manuscript
• Wikipedia information flow analysis reveals the scale-free architecture of the Semantic Space
• Good practices for a literature survey are not followed by authors while preparing scientific manuscripts
• Word sense disambiguation via high order of learning in complex networks
• Authorship recognition via fluctuation analysis of network topology and word intermittency
• Three-feature model to reproduce the topology of citation networks and the effects from authors' visibility on their h-index
• Identification of Literary Movements Using Complex Networks to Represent Texts
• Comparing intermittency and network measurements of words and their dependency on authorship
• On time-varying collaboration networks
• Complex networks analysis of language complexity
• Extracting information from S-curves of language change
• Modeling the average shortest path length in growth of word-adjacency networks
• A stronger null hypothesis for crossing dependencies
• Extracting directed information flow networks: an application to genetics and semantics
• Modeling the Co-occurrence Principles of the Consonant Inventories: A Complex Network Approach

Cited by in corpus (12)

• Text-mining forma mentis networks reconstruct public perception of the STEM gender gap in social media
• Using word embeddings to improve the discriminability of co-occurrence text networks
• Word sense induction using word embeddings and community detection in complex networks
• Linguistic data mining with complex networks: a stylometric-oriented approach
• On the role of words in the network structure of texts: application to authorship attribution
• Writing about COVID-19 vaccines: Emotional profiling unravels how mainstream and alternative press framed AstraZeneca, Pfizer and vaccination campaigns
• Semantic flow in language networks
• Punctuation patterns in "Finnegans Wake" by James Joyce are largely translation-invariant
• Universal and non-universal text statistics: Clustering coefficient for language identification
• Language Networks: a Practical Approach
• Average shortest-path length in word-adjacency networks: Chinese versus English
• Cognitive network science quantifies feelings expressed in suicide letters and Reddit mental health communities