A hybrid lattice-based commitment scheme is proposed for anonymous proofs between hidden values. The method is based on a modification of the BDLOP zero-knowledge proof (ZKP) scheme by replacing the learning with errors (LWE) problem with a learning with rounding (LWR) problem, which theoretically makes it possible to reduce the size of the parameters and reduce the complexity of parameter selection. It is shown that the proposed scheme preserves the property of additive homomorphism, which makes it possible to apply it to prove linear relations. The obtained results can be applied to construct electronic voting protocols or conduct anonymous transactions.

This article presents methods for the automated security analysis of Android applications, which can be used to search for cryptographic vulnerabilities, vulnerabilities in third-party software components, authentication, and authorization, as well as to identify the storage and transmission of sensitive information in clear text. The accuracy of searching for the given types of vulnerabilities using automated vulnerability search tools and a software prototype is analyzed.

Existing methods for detecting attacks on data integrity in file systems are studied. A detection method based on the use of several entropy metrics is proposed. The effectiveness of the proposed method is evaluated using the example of detecting existing ransomware programs.

Immunization approaches for cyber-physical systems based on hierarchical and peer-to-peer network topologies are described. A systematic analysis of existing immunization methods based on graph theory is carried out, together with theoretical formalization and experimentation. The approach demonstrates sufficient effectiveness of artificial immunization using global strategies for cyber-physical systems based on a hierarchical network infrastructure.

One of the key elements in solving spam filtering problems is the text vectorization method. This article proposes a vectorization method based on matching text to pairs of intentionalities. A list of intentionality pairs is extracted and a synthetic dataset is generated from text utterances. A neural network is designed and trained to determine the degree to which each intentionality belongs to the textual expression provided as the input of the model. The developed method is tested on the problem of filtering spam messages using logistic regression and the Enron dataset and SMS dataset.

The principles of construction and operation of honeypot systems are studied. Existing detection methods are analyzed, and their advantages and disadvantages are highlighted. A detection method based on the analysis of command execution delays is proposed. A universal detection method based on combining the results of the methods is proposed. A software prototype of the detection system is developed and tested.

The problem of ensuring the security of a global computational model in federated learning systems is considered. A method is proposed that is based on data verification using a trusted group of nodes and ensures that only correct updates are taken into account during the global model aggregation process. It is experimentally demonstrated that the developed method ensures accurate identification and isolation of adversaries implementing label-flipping and noise-injection threats.

This paper studies the detection of intrusions and breaches of data confidentiality stored in database management systems (DBMSs) based on behavioral analysis. A major challenge in this domain lies in considering not only the syntax of the query but also the semantic relationships of the data, since syntactic and contextual approaches fail to detect all types of attacks. Based on an analysis of well-known studies, a method for detecting anomalies in user behavior is proposed based on the author’s behavior assessment metrics and the scope of the requested data. The proposed method develops a well-known work, while significantly improving the detection of certain types of behavioral deviations. An important part of this study involves identifying the features of the application of this type of analysis and its limitations.

The problem of detecting unreliable news content is studied and a solution based on machine learning methods is proposed. Modern approaches used to assess the reliability of textual and multimedia content are analyzed, with promising approaches identified and adapted to the Russian-language media space. A combined method for detecting fake news is proposed, based on the joint analysis of textual and multimedia information, as well as the characteristics of content dissemination. Testing the proposed method confirmed its effectiveness and applicability for the automated detection of unreliable news content in real-world information systems.

A method for detecting and eliminating latent vulnerabilities in mobile applications at the final stage of development is presented. The nature of such vulnerabilities is analyzed and existing methods for their identification and neutralization are considered. The developed method is based on the analysis of the application’s state at different points in time and comparison with the reference state. Includes recording the initial state, discrete analysis of its changes, detection and identification of vulnerabilities, and their elimination. The main area of the application of the method is to ensure the security of mobile applications in the late stages of development and during final testing. The findings of the study highlight the importance of early detection of latent vulnerabilities to ensure a high level of information security of the final software product.