Journal of Hacking Techniques, Digital Crime Prevention and Computer Virology

Cyber Risk Governance Framework for Multi-Cloud Environments: An Empirical Study of Nigerian Organisations

2026-06-06T04:31:11+00:00

The rapid adoption of multi-cloud environments, where organisations simultaneously use two or more cloud service providers, has introduced a complex and layered cyber risk landscape that existing single-cloud governance frameworks are not equipped to manage. This is particularly acute in Nigeria, where banking, education, healthcare, and government institutions are increasingly dependent on cloud infrastructure yet lack a unified governance approach that applies consistently across all providers in use. This study addresses that gap by developing and validating the Cyber Risk Governance Framework for Multi-Cloud Environments (CRGF-MCE), a structured, empirically grounded, and mathematically validated governance model. A descriptive survey design was adopted, and a structured 33-item questionnaire was administered to 100 respondents drawn from Nigerian organisations operating multi-cloud environments across multiple sectors. Three differentiated Likert-type scales were used across five thematic sections covering cloud adoption, cyber threat monitoring, governance structures, risk controls, and governance challenges. A one-sample t-test was applied to test the hypothesis that significant cyber threats exist in multi-cloud environments. The hypothesis test produced a t-statistic of 17.09, far exceeding the critical value of 1.645 at α = 0.05, confirming the significant presence of seven threat categories, including misconfigurations, IAM weaknesses, insider threats, and unified visibility gaps. The CRGF-MCE was developed, comprising five integrated components and four governance layers, namely Identity and Access Management, Logging and Monitoring, Compliance Enforcement, and Policy Orchestration and was validated with a Model Validation Score of 0.7743. The framework provides Nigerian organisations with a replicable, scalable, and continuously adaptive governance system for multi-cloud environments.

Empirical Evaluation Model for Differential Privacy in Federated Learning for Cyber Threat Intelligence Sharing

2026-07-30T09:04:29+00:00

Today, the threat landscape is evolving rapidly, with many organizations continually facing complex and malicious cyber threats. Collective Cyber Threat Intelligence (CTI) sharing has become a proactive approach for addressing cyber threats. However, there is a drawback in sharing raw CTI data due to privacy and confidentiality concerns. Although Federated Learning (FL) remains vulnerable to privacy attacks, Differential Privacy (DP) has become the de facto standard for privacy. However, there is a critical gap in how DP affects utility and the extent of privacy loss in practice. In this study, a four-layer FL model that partitions the class-imbalanced CICIDS-2017 dataset across five simulated organizations was designed. Local parameters of 3-layer Multilayer Perceptron classifiers are aggregated over ten communication rounds using a custom weight-persisting strategy; the utility of the global model was then evaluated using Macro-F1, and its empirical privacy was audited using black-box loss-threshold attacks, shadow classifiers, and reference-model Likelihood Ratio Audits (LiRA). A Heterogeneous Differential Privacy Federated Learning mechanism (HetDP-FL) that proposes and validates the selection of a client-level differential privacy budget based on the risk tolerance of each participating organization was implemented. The results of HetDP-FL improve the Macro-F1 score by +7.3% when compared to the uniform privacy budget. Results show that the benefits of federated CTI systems can be achieved by integrating formal DP guarantees with empirical privacy auditing for a particular deployment and at participant-level privacy heterogeneity.

Attention-Based LSTM and Autoencoder Framework for Credit Card Fraud Detection Using SMOTE

2026-07-03T09:36:47+00:00

The increase in the adoption of credit cards, especially through online channels, has caused an upsurge in the number of fraudulent activities. This has caused substantial losses to both financial institutions and their clients. Hence, there is a need for the development of efficient credit card fraud detection systems that can accurately detect any potential threats. The current paper suggests a new fraud detection system that combines two different methodologies. It uses autoencoders to carry out anomaly detection alongside a deep learning model that utilizes an attention-based long-short-term memory network. As opposed to the classical approach, which views each transaction as an independent unit, the current model takes into account the sequential nature of the transaction data, allowing it to recognize patterns and behavior of fraudsters. The synthetic minority oversampling technique is used during data preparation to solve the challenge of class imbalance prevalent in fraud datasets. Also, the attention mechanism will be implemented in the LSTM architecture to identify the important transactions involved in the process of fraud detection. The performance of the proposed architecture is analyzed and compared against that of existing machine learning algorithms.

Low Power Artificial Intelligence on Edge Devices: Implementation on Raspberry Pi and ESP32, an Ultra-Low Power Microcontroller for Edge AI Applications

2026-07-28T10:18:47+00:00

Low-power AI on edge devices is something that is getting a lot of attention lately. It lets systems handle data right where it is, on the device itself, instead of always sending everything to the cloud. That cuts down on delays, keeps things more private, and uses less network and power overall. That makes it pretty useful for health monitoring or smart homes, where quick responses are required. The goal of this study is to build and test simple AI models that perform well on limited hardware, without sacrificing much accuracy or slowing down. Hardware selection is the initial step, based on processing capability, memory capacity, power consumption, and connectivity requirements. Raspberry Pi is good for heavier tasks, like spotting objects in images or video stuff, since it has more power and RAM, plus it runs libraries easily.

On the other hand, ESP32 is for lighter jobs, say watching sensors or recognizing voice commands, because it consumes less power and has built-in WiFi and Bluetooth. For implementation, data are collected from cameras or sensors, depending on the application. The data are then preprocessed by cleaning, resizing, and splitting them into training and testing datasets. This process improves the performance and generalization capability of the models. The models are built in Python with TensorFlow Lite, using things like basic CNNs or TinyML that do not need many resources. To improve computational efficiency, optimization techniques such as quantization and pruning are applied to reduce model size and computational requirements. Deployment strategies vary depending on the target hardware platform and application requirements. On Raspberry Pi, OpenCV and TensorFlow are used for images, while ESP32 uses Arduino or Embedded C for simpler embedded work. Evaluating means checking accuracy, how long inference takes, CPU and memory use, and power draw. Results show Raspberry Pi handles complex tasks with higher accuracy, but ESP32 shines when power and portability matter more, like for always-on devices. Both platforms work for edge AI; it depends on the app and power available. Some applications might need one over the other, and that choice is not always straightforward.