Digital video surveillance systems are now common in the security infrastructure of modern times, but proprietary file systems provided by large manufacturers are a major challenge to the work of the forensic investigator. This paper proposes a forensic recovery methodology of Hikvision and Dahua surveillance systems by utilizing three major innovations: (1) adaptive temporal sequencing, which dynamically changes gap detection thresholds; (2) dual-signature validation with header–footer matching of DHFS frames; and (3) automatic manufacturer identification. The strategy puts into practice direct binary analysis of proprietary file systems, frame-based parsing and automatic video reconstruction. Testing on 27 surveillance hard drives showed a recovery rate of 91.8, a temporal accuracy of 96.7% and a false positive rate of 2.4%—the lowest of the tools tested with statistically significant improvements over commercial tools (p < 0.01). Better results with fragmented streams (87.2 vs. 82.4% with commercial tools) meet key forensic needs of determining valid evidence chronology. The open methodology offers the necessary algorithmic transparency to be court-admissible, and the automated MP4 conversion with metadata left intact makes the integration of forensic workflow possible. The study provides a scientifically validated approach to proprietary surveillance formats, which evidences technical innovativeness and practical usefulness to digital forensics investigations.

According to the latest Verizon DBIR report, credential abuse, including password reuse and human factors in password creation, remains the leading attack vector. It was revealed that most users change their passwords only when they forget them, and 35% of respondents find mandatory password rotation policies inconvenient. These findings highlight the importance of combining technical solutions with user-focused education to strengthen password security. In this research, the “human factor in the creation of usernames and passwords” is considered a vulnerability, as identifying the patterns or rules used by users in password generation can significantly reduce the number of possible combinations that attackers need to try in order to gain access to personal data. The proposed method based on an LSTM model operates at a character level, detecting recurrent structures and generating generalized masks that reflect the most common components in password creation. Open datasets of 31,000 compromised passwords from real-world leaks were used to train the model and it achieved over 90% test accuracy without signs of overfitting. A new method of evaluating the individual password creation habits of users and automatically fetching context-rich keywords from a user’s public web and social media footprint via a keyword-extraction algorithm is developed, and this approach is incorporated into a web application that allows clients to locally fine-tune an LSTM model locally, run it through ONNX, and carry out all inference on-device, ensuring complete data confidentiality and adherence to privacy regulations.

Predicting Local Area Network (LAN) equipment failure is of utmost importance to ensure the uninterrupted operation of modern communication networks. This study explores the use of machine learning algorithms to enhance the accuracy of equipment failure prediction in LAN environments. Using these algorithms to enhance LAN failure predictions involves collecting and analyzing network data, such as packet loss rates and latency, to identify patterns and anomalies. These algorithms can then predict potential LAN failures by recognizing early warning signs and deviations from normal network behavior. By leveraging machine learning, network administrators can proactively address issues, reduce downtime, and improve overall network reliability. In our study, two powerful machine learning algorithms—decision tree and support vector machine (SVM)—are used. To evaluate the effectiveness of the proposed models, a comprehensive dataset comprising various LAN equipment parameters and corresponding failure instances is utilized. The dataset is pre-processed to handle missing values and normalize features, ensuring the algorithms’ optimal performance. Performance metrics, such as accuracy, precision, recall, and F1-score, are employed to assess the predictive capabilities of the models. The excremental results of our study lead to more reliable and stable network operations by allowing early detection of potential issues and preventive maintenance. This leads to reduced downtime, improved network performance, and enhanced overall user satisfaction. They demonstrate the efficacy of both decision tree and SVM algorithms in accurately predicting LAN equipment failure.

This research develops a method for synthesizing linguistic models of fuzzy systems with fuzzy input and output parameters that are described by linguistic variables. Based on the proposed method, linguistic models of the Title 1000 catalytic cracking unit for heavy residues at the Shymkent oil refinery are developed, describing the dependence of the volume and quality of gasoline on the input and operating parameters of the facility, which are fuzzy. It is substantiated that the use of a fuzzy approach, which allows the use of the experience, knowledge, and intuition (intelligence) of the decision maker and subject matter experts, is the most suitable effective method for synthesizing models of complex, fuzzily described objects and processes for comparison with other methods. The main idea of the proposed work is to solve the problems of shortage and fuzziness of initial information when developing models and optimizing the operating modes of a catalytic cracking unit through the use of knowledge, experience, and intuition of experts in this field. To solve the problems of the shortage of initial quantitative information and the fuzziness of available information when developing mathematical models, it is proposed to systematically use statistical methods, expert assessment methods, and a heuristic method based on fuzzy logic. The scientific novelty of the research lies in the development of a method for synthesizing linguistic models in a fuzzy environment and an algorithm for its implementation, which makes it possible to describe the dependence of the fuzzy values of the object’s output parameters on its fuzzy input and operating parameters. The proposed approach allows the formalization and synthesis of models of fuzzily described objects when other methods of model development are not applicable or do not give the expected results. The results of the work were simulated in the MATLAB Fuzzy Logic Toolbox.

This article proposes a new concept of a smart office that allows the monitoring of employees' health. According to the literature review, the technology used to monitor office workers' health is still in research. The statistics prove the relevance of the human health monitoring system. The paper proposes the smart office concept that will cover the designed structural model of the health monitoring system. The principles and criteria that determine the need to introduce such a system to the workplaces of office employees are defined. The physical model demonstrating the content of the system is described. The diagram of the data flow is presented. The structural scheme of the algorithm is developed. A description of the database and server is given. The mathematical foundation of the deep learning system is described.