Authorship identification of documents with high content similarity

Abstract

The goal of our work is inspired by the task of associating segments of text to their real authors. In this work, we focus on analyzing the way humans judge different writing styles. This analysis can help to better understand this process and to thus simulate/ mimic such behavior accordingly. Unlike the majority of the work done in this field (i.e. authorship attribution, plagiarism detection, etc.) which uses content features, we focus only on the stylometric, i.e. content-agnostic, characteristics of authors. Therefore, we conducted two pilot studies to determine, if humans can identify authorship among documents with high content similarity. The first was a quantitative experiment involving crowd-sourcing, while the second was a qualitative one executed by the authors of this paper. Both studies confirmed that this task is quite challenging. To gain a better understanding of how humans tackle such a problem, we conducted an exploratory data analysis on the results of the studies. In the first experiment, we compared the decisions against content features and stylometric features. While in the second, the evaluators described the process and the features on which their judgment was based. The findings of our detailed analysis could (1) help to improve algorithms such as automatic authorship attribution as well as plagiarism detection, (2) assist forensic experts or linguists to create profiles of writers, (3) support intelligence applications to analyze aggressive and threatening messages and (4) help editor conformity by adhering to, for instance, journal specific writing style.

Towards a Stylometric Authorship Recognition Model for the Social Media Texts in Arabic

“…The assumption is that documents or texts clustered together are more likely to be written by the same author. Rexha, Kröll, Ziak, and Kern (2018) explain that authorship recognition can be done using document clustering where the author of a disputed or controversial text can be identified from a set of candidate authors. Theodoridis and Koutroubas (2003) suggest that text clustering is one of the most primitive mental activities of humans.…”

Interpol review of questioned documents 2016–2019

“…The text analysis is field with different topic as the linguistic [ [206] , [207] , [208] ], the stylometry [ 209 ], and text classification [ 210 ].…”

Analysis of Stylometric Features and Segmentation Strategies in Intrinsic Plagiarism Detection System

“…Seiring dengan proses otomatisasi di segala bidang, maka makna stilometri mengalami pergeseran dan difenisikan oleh Halvani [1] sebagai "cabang ilmu yang menentukan kepemilikan pengarang terhadap karya-karya tulis melalui analisis statistik dan pembelajaran mesin". Analisis stilometri banyak diterapkan dalam aplikasi komputasional yang lebih kompleks seperti pada Identifikasi Kepengarangan (authorship identification) oleh Rexha dkk [2], atribusi dan diarisasi penulis oleh Stamatatos dkk [3], atau Deteksi Plagiasi Intrinsik (DPI) oleh Rexha [1] dan Kuznetsov dkk [4].…”

“…Selain itu, sebagian besar sistem DPI menggunakan beberapa fitur sekaligus daripada hanya mengandalkan fitur tunggal. Fitur stilometri lainnya yang kerap digunakan adalah frekuensi panjang kata [2], [12], Frekuensi panjang kalimat [12], [13], Frekuensi tag kelas kata (part of speech (pos) tag frequency) [12], Rasio type-token [2], [13], dan frekuensi kespesifikan kata [12].…”