Anna Kurtukova

With the development of digital interaction, there is a need to control the information distributed by users so that it complies with legal requirements. The study proposes a technique for classifying Russian-language texts by considering both the biological sex of the authors (male and female) and gender-related distinctions, including heterosexual and homosexual groups as well as specific LGBT categories such as gay men, lesbian women, bisexual, and transgender authors. The article proposes a technique for identifying the sex and gender of authors of Russian-language texts by utilizing an ensemble of methods, including Support Vector Machines (SVM), Convolutional Neural Networks (CNN), and the Russian Bidirectional Encoder Representations from Transformers (RuBERT). This research is introducing a technique for gender identification in Russian texts, as prior studies have only addressed the automatic determination of an author’s sex. To conduct the study, custom datasets of user comments from social networks were created and annotated with both sex and gender labels. In tasks related to identifying the author’s sex, the proposed technique achieved an accuracy exceeding 90 %. This includes the classification of heterosexual men and women as well as a mixed dataset comprising both heterosexual and homosexual individuals. When distinguishing individuals of the same sex based on sexual orientation, the results showed that homosexual and heterosexual women display uncommon writing styles in contrast to men of different orientations, with accuracy rates of 93 % for women and 85 % for men. Additionally, an experiment focused on identifying LGBT individuals and their gender identities based on writing style achieved an accuracy of 93 %. The key takeaway from this study is that combining CNN, RuBERT, and SVM leads to a more robust model for gender classification. The method has been tested on varying numbers of samples. In the basic experiments for determining sex, the method achieves an accuracy of 92 %. When addressing the task of detecting the author’s LGBT affiliation, the method achieves 93 % accuracy. The proposed technique can be applied for automated monitoring of social networks to detect and analyze the gender identity of text authors. It is also promising for use in content moderation systems as well as in sociological and linguistic research.

The issue of text plagiarism in academic and educational environments is becoming increasingly relevant every year. The quality of research articles and works is declining due to students copying fragments of others’ works and using modern generative models for text and source code creation. The article proposes an integrated technique for authorship verification of both natural and programming language texts, based on a combination of statistical methods, machine learning, and deep neural networks. The presented technique addresses several related tasks: assessing text homogeneity, detecting plagiarism when solving closed-set authorship attribution problems, and identifying texts and fragments created by generative models. Experimental data include a multi-domain dataset of natural language texts consisting of research articles on natural and technical sciences, PhD dissertations, and artificially generated samples on related topics. To evaluate the effectiveness of the technique in relation to programming language texts, a multilingual program dataset was used, consisting of source codes for programs of technical students as well as artificially generated program codes. The experimental results demonstrate the effectiveness of the proposed technique for plagiarism detection and copyright protection in the educational process. The accuracy of identifying heterogeneous fragments in text or code is 93-94%, authorship attribution ac-curacy is 89-99% depending on the number of co-authors, and verification accuracy is 97.5-99.4%.

This paper is the fourth part of a research series that focuses on determining the authorship of Russian-language texts by analyzing short social media comments, including those from mass media and communities associated with destructive content. Semantic text clustering was used to analyze content and employed a transfer learning technique based on a pre-trained model to identify sensitive topics. Authorship attribution is implemented as a classical classification task with a closed set of authors and a more challenging open-set task. In the latter case, multiple experiments were conducted, incorporating the identification of destructive content with known authors and artificially generated texts. For open attribution, a method combining One-Class SVM and fastText was proposed. Results demonstrate high accuracy (92% or higher) for cases with 2 and 5 authors, regardless of comment length and the additional task of identifying authors of destructive text. Mixed-data experiments involving 10 or more authors yielded results comparable to or more accurate (84% or higher) than previous studies.

This article is part of a series aimed at determining the authorship of source codes. Analyzing binary code is a crucial aspect of cybersecurity, software development, and computer forensics, particularly in identifying malware authors. Any program is machine code, which can be disassembled using specialized tools and analyzed for authorship identification, similar to natural language text using Natural Language Processing methods. We propose an ensemble of fastText, support vector machine (SVM), and the authors’ hybrid neural network developed in previous works in this research. The improved methodology was evaluated using a dataset of source codes written in C and C++ languages collected from GitHub and Google Code Jam. The collected source codes were compiled into executable programs and then disassembled using reverse engineering tools. The average accuracy of author identification for disassembled codes using the improved methodology exceeds 0.90. Additionally, the methodology was tested on the source codes, achieving an average accuracy of 0.96 in simple cases and over 0.85 in complex cases. These results validate the effectiveness of the developed methodology and its applicability to solving cybersecurity challenges.

This article is the third paper in a series aimed at the establishment of the authorship of Russian-language texts. This paper considers methods for determining the authorship of classical Russian literary texts, as well as fanfiction texts. The process of determining the author was first considered in the classical version of classification experiments using a closed set of authors, and experiments were also completed for a complicated modification of the problem using an open set of authors. The use of methods to identify the author of the text is justified by the conclusions about the effectiveness of the fastText and Support Vector Machine (SVM) methods with the selection of informative features discussed in our past studies. In the case of open attribution, the proposed methods are based on the author’s combination of fastText and One-Class SVM as well as statistical estimates of a vector’s similarity measures. The feature selection algorithm for a closed set of authors is chosen based on a comparison of five different selection methods, including the previously considered genetic algorithm as a baseline. The regularization-based algorithm (RbFS) was found to be the most efficient method, while methods based on a complete enumeration (FFS and SFS) are found to be ineffective for any set of authors. The accuracy of the RbFS and SVM methods in the case of classical literary texts averaged 83%, which outperforms other selection methods by 3 to 10% for an identical number of features, and the average accuracy of fastText was 84%. For the open attribution in cross-topic classification, the average accuracy of the method based on the combination of One-Class SVM with RbFS and fastText was 85%, and for in-group classification, it was 75 to 78%, depending on the group, which is the best result among the open attribution methods considered.

This paper is a continuation of our previous work on solving source code authorship identification problems. The analysis of heterogeneous source code is a relevant issue for copyright protection in commercial software development. This is related to the specificity of development processes and the usage of collaborative development tools (version control systems). As a result, there are source codes written according to different programming standards by a team of programmers with different skill levels. Another application field is information security—in particular, identifying the author of computer viruses. We apply our technique based on a hybrid of Inception-v1 and Bidirectional Gated Recurrent Units architectures on heterogeneous source codes and consider the most common commercial development complex cases that negatively affect the authorship identification process. The paper is devoted to the possibilities and limitations of the author’s technique in various complex cases. For situations where a programmer was proficient in two programming languages, the average accuracy was 87%; for proficiency in three or more—76%. For the artificially generated source code case, the average accuracy was 81.5%. Finally, the average accuracy for source codes generated from commits was 84%. The comparison with state-of-the-art approaches showed that the proposed method has no full-functionality analogs covering actual practical cases.

Статья посвящена идентификации автора неоднородного исходного кода программы на основе гибридной нейронной сети. Решения данной проблемы особенно актуальны для областей информационной безопасности, образовательного процесса и защиты авторского права. В статье представлен анализ современных методов решения поставленной задачи. Авторами предлагается собственная методика на основе гибридной нейронной сети, зарекомендовавшей себя в ранних исследованиях, направленных на оценку эффективности данного подхода в простых и сложных случаях. Данная работа включает в себя эксперименты по ранее не рассмотренным случаям идентификации автора исходного кода на основе неоднородных данных. Рассматриваются случаи, актуальные для корпоративной разработки. Среди них анализ исходных кодов, представленных в виде коммитов, и обучение модели на наборах данных, включающих в себя два и более языка программирования. Также исследовано набирающее популярность направление определения авторства искусственно сгенерированного исходного кода. Для каждого случая сформирован набор данных и проведен эксперимент. Оценка эффективности авторской методики для всех трех сложных случаев осуществлена при помощи перекрестной проверки по 10 блокам. Средняя точность для смешанных наборов данных составила 87 % для двух языков программирования и 76 % для трех и более языков, соответственно. Точность методики для решения задачи определения авторства искусственно сгенерированных исходных кодов в среднем составила 81,5 %. Идентификация автора исходного кода программы на основе коммитов осуществлялась с точностью 84 %. Эксперименты показали, что во всех трех случаях эффективность методики может быть повышена путем использования больших объемов обучающих данных. Ключевые слова: авторство, исходный код, коммиты, генерация, нейронная сеть. The article is devoted to the issue of identifying the author of a heterogeneous source code program by means of a hybrid neural network. The solutions to this problem are especially relevant to the fields of information security, educational process, and copyright protection. The article analyzes modern methods of addressing this problem. The authors propose their own methodology based on a proven in early studies hybrid neural network aimed at evaluating the effectiveness of this approach in simple and difficult cases. This research incorporates experiments on previously unconsidered cases of source code author identification based on heterogeneous data. Cases relevant to corporate development are examined including the analysis of source codes presented as commits and model training on datasets with more than two programming languages. Additionally, the trend of determining the authorship of an artificially generated source code, which is gaining traction, is regarded. A dataset was generated, and an appropriate experiment was performed for each case. The effectiveness of the author's methodology for all three difficult cases was evaluated using a 10 blocks cross-validation. The average accuracy for mixed datasets was 87 % for two programming languages and 76 % for three or more languages, respectively. The average accuracy of the methodology for authorship identification of artificially generated source codes was 81.5 %. Identification of the author of a program source code based on commits was carried out with an accuracy of 84 %. Experiments have shown that the effectiveness of the methodology can be improved in all three cases by using large amounts of training data.

Authorship attribution is one of the important fields of natural language processing (NLP). Its popularity is due to the relevance of implementing solutions for information security, as well as copyright protection, various linguistic studies, in particular, researches of social networks. The article is a continuation of the series of studies aimed at the identification of the Russian-language text’s author and reducing the required text volume. The focus of the study was aimed at the attribution of textual data created as a product of human online activity. The effectiveness of the models was evaluated on the two Russian-language datasets: literary texts and short comments from users of social networks. Classical machine learning (ML) algorithms, popular neural networks (NN) architectures, and their hybrids, including convolutional neural network (CNN), networks with long short-term memory (LSTM), Bidirectional Encoder Representations from Transformers (BERT), and fastText, that have not been used in previous studies, were applied to solve the problem. A particular experiment was devoted to the selection of informative features using genetic algorithms (GA) and evaluation of the classifier trained on the optimal feature space. Using fastText or a combination of support vector machine (SVM) with GA reduced the time costs by half in comparison with deep NNs with comparable accuracy. The average accuracy for literary texts was 80.4% using SVM combined with GA, 82.3% using deep NNs, and 82.1% using fastText. For social media comments, results were 66.3%, 73.2%, and 68.1%, respectively.

Reconstructed 3D foot models can be used for 3D printing and further manufacturing of individual orthopedic shoes, as well as in medical research and for online shoe shopping. This study presents a technique based on the approach and algorithms of photogrammetry. The presented technique was used to reconstruct a 3D model of the foot shape, including the lower arch, using smartphone images. The technique is based on modern computer vision and artificial intelligence algorithms designed for image processing, obtaining sparse and dense point clouds, depth maps, and a final 3D model. For the segmentation of foot images, the Mask R-CNN neural network was used, which was trained on foot data from a set of 40 people. The obtained accuracy was 97.88%. The result of the study was a high-quality reconstructed 3D model. The standard deviation of linear indicators in length and width was 0.95 mm, with an average creation time of 1 min 35 s recorded. Integration of this technique into the business models of orthopedic enterprises, Internet stores, and medical organizations will allow basic manufacturing and shoe-fitting services to be carried out and will help medical research to be performed via the Internet.

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.