Authorship Identification of a Russian-Language Text Using Support Vector Machine and Deep Neural Networks

Abstract

The article explores approaches to determining the author of a natural language text and the advantages and disadvantages of these approaches. The importance of the considered problem is due to the active digitalization of society and reassignment of most parts of the life activities online. Text authorship methods are particularly useful for information security and forensics. For example, such methods can be used to identify authors of suicide notes, and other texts are subjected to forensic examinations. Another area of application is plagiarism detection. Plagiarism detection is a relevant issue both for the field of intellectual property protection in the digital space and for the educational process. The article describes identifying the author of the Russian-language text using support vector machine (SVM) and deep neural network architectures (long short-term memory (LSTM), convolutional neural networks (CNN) with attention, Transformer). The results show that all the considered algorithms are suitable for solving the authorship identification problem, but SVM shows the best accuracy. The average accuracy of SVM reaches 96%. This is due to thoroughly chosen parameters and feature space, which includes statistical and semantic features (including those extracted as a result of an aspect analysis). Deep neural networks are inferior to SVM in accuracy and reach only 93%. The study also includes an evaluation of the impact of attacks on the method on models’ accuracy. Experiments show that the SVM-based methods are unstable to deliberate text anonymization. In comparison, the loss in accuracy of deep neural networks does not exceed 20%. Transformer architecture is the most effective for anonymized texts and allows 81% accuracy to be achieved.

Authorship Attribution of Social Media and Literary Russian-Language Texts Using Machine Learning Methods and Feature Selection

“…Practice shows that using simpler methods proven is more justified than novel approaches in different cases. As such, in our last work [1] SVM's accuracy was comparable to the more modern methods of deep learning, while SVM was training much faster. Therefore, it was decided to expand the list of classical methods and test SVM, LR, NB, DT, RF, KNN in authorship identification.…”

“…Numerous studies have proven the ability of SVM in identifying the authorship of literary texts [1,6,41]. ML algorithms most often require structured data, while deep NN are capable of analyzing text sequences and selecting informative features automatically.…”

“…There are many studies devoted to establishing authorship of a natural text and source code, the gender and age of the author, determining the sentiment of the text for forensic and social science purposes [5,6]. The early paper [1] provides a detailed review of 2015-2020 studies aimed at determining the author of a text, including approaches based on deep neural networks (NN), classical machine learning (ML) methods, and aspect analysis. This section examines the novel studies for Russian and other languages, as well as the works devoted to feature selection.…”

“…Everyone has an opportunity to express his opinions, share thoughts, and receive feedback from followers. The variety of available online texts (e.g., emails, blogs, correspondence in social networks, and messengers) and possibility to write them anonymously indicate a wide range of applications of methods for determining the author of the text [1]. The number of search results of the query "Authorship attribution" in Google Scholar for 2021 is 23,000, for 2017-2021-60,000.…”

Data Science and Knowledge Discovery

“…The eleventh paper, entitled Authorship Identification of a Russian-Language Text Using Support Vector Machine and Deep Neural Networks was written by Aleksandr Romanov, Anna Kurtukova, Alexander Shelupanov, Anastasia Fedotova, and Valery Goncharov [11]. The authors explored the advantages and disadvantages of various approaches that can determine the author of a natural language text.…”