This paper introduces the Gray Wolf Optimized Convolutional Transformer Network, a combined deep learning framework aimed at accurately and efficiently recognizing dynamic hand gestures, especially in American Sign Language (ASL). The model integrates Convolutional Neural Networks (CNNs) for spatial feature extraction, Transformers for temporal sequence modeling, and Grey Wolf Optimization (GWO) for hyperparameter tuning. Extensive experiments were conducted on two benchmark datasets, ASL Alphabet and ASL MNIST to validate the model’s effectiveness in both static and dynamic sign classification. The proposed model achieved superior performance across all key metrics, including a accuracy of 99.40%, F1-score of 99.31%, Matthews Correlation Coefficient (MCC) of 0.988, and Area Under the Curve (AUC) of 0.992, surpassing existing models such as PCA-IGWO, KPCA-IGWO, GWO-CNN, and AEGWO-NET. Real-time gesture detection outputs further demonstrated the model’s robustness in varied environmental conditions and its applicability in assistive communication technologies. Additionally, the integration of GWO not only accelerated convergence but also enhanced generalization by optimally selecting model configurations. The results show that GWO-CTransNet offers a powerful, scalable solution for vision-based sign language recognition systems, combining high accuracy, fast inference, and adaptability in real-world applications. 

The continued rise of the metaverse ecosystems and Augmented Reality (AR) technologies has shown novel dimensions of digital interaction, however, it has also created novel cybersecurity and forensics related challenges. The traditional Digital Forensic (DF) techniques are seen to struggle to address the dynamic and immersive nature of AR environments. This, has been seen in situations, where cybersecurity incidents can be complex and hard to be understood. This paper, therefore, presents a first step toward a conceptual Metaverse Forensics Model (MFM) that proactively bridges AR and potential cybesecurity-related incidents, offering a structured approach to a Digital Forensic Investigation (DFI) in virtual and hybrid spaces. To ensure compliance the suggested processes are aligned to ISO/IEC 27043 International Standard. The MFM suggests a possible integration of real-time monitoring, Potential Digital Evidence (PDE) preservation in a Digital Forensic Readiness (DFR) approach as a proactive step towards a DFI. In this context, DFR is being integrated due to the costly nature of a DFI and the unnecessary requirement of changing or tampering with the meteverse infrastructure during a DFI. On the same note, the proposed MFM model enables law enforcement, cybersecurity professionals, and DF investigators to strengthen digital attribution in metaverse ecosystems. An evaluation on the posible feasibility and effectiveness of the MFM has been conducted through a comparative analysis with the most-recent studies based on selected metrics, and from the outcome it has been seen that this area is still at infancy stage, however, the conclusion shows the need for standards and adaptive DF methodologies. 

Collaborative AI engineering is a paradigm that enables multiple stakeholders to maintain AI pipelines by exchanging artifacts, such as data, models, and software packages. It is a cost-efficient engineering process that accelerates the development of AI applications, specifically for small and medium-sized enterprises (SMEs). However, AI artifacts are often associated with inherent intellectual property (IP) or sensitive information, which hinders collaboration due to a lack of trust among stakeholders. Digital sovereignty is viewed as an ideal state where artifact owners maintain full control over their assets, such as making key decisions regarding access, storage, and interoperability. Thus, it is postulated that implementing digital sovereignty mechanisms in multi-stakeholder information systems can address the trust gap and promote collaboration. This work addresses this gap by conducting a systematic review of the literature on digital sovereignty for collaborative AI engineering. The main contributions of this work include: i) mapping digital sovereignty definitions and requirements within the context of collaborative AI engineering, ii) identifying existing technologies and concepts for implementing sovereignty features in AI engineering, and iii) analyzing existing collaborative AI platforms such as data marketplaces, data spaces, and GAIA-X. This analysis highlights their sovereignty requirements, solutions, benefits, and implementation challenges. In addition, this work iv) identifies research gaps in data pricing, confidentiality, and interoperability, and proposes future directions to enhance digital sovereignty in collaborative AI ecosystems. 

Digital Forensics (DF) in distributed environments faces significant challenges, ranging from scalability, complexity, and reliance on traditional DF processes. The problem being addressed in this paper, is the lack of effective automated DF analysis across distributed ecosystems. Inspired by the success of peer-to-peer (P2P) architectures, and as a step toward overcoming the limitations of traditional client-server models, a Distributed Adaptive Push-button Forensic (DAPF) System that leverages a decentralized approach is suggested. The DAPF system automates attack data collection and analysis across multiple nodes in an adaptive approach to streamline DF investigations. Preliminary experiments have demonstrated a 30% reduction in analysis time compared to traditional methods. This work highlights the potential of automation, adaptability, and decentralized architectures in modern DF a step towards distributed digital forensics. 

Digital Forensics (DF) in distributed environments faces significant challenges, ranging from scalability, complexity, and reliance on traditional DF processes. The problem being addressed in this paper, is the lack of effective automated DF analysis across distributed ecosystems. Inspired by the success of peer-to-peer (P2P) architectures, and as a step toward overcoming the limitations of traditional client-server models,  a Distributed Adaptive Push-button Forensic (DAPF) System that leverages a decentralized approach is suggested. The DAPF system automates attack data collection and analysis across multiple nodes in an adaptive approach  to streamline DF investigations. Preliminary experiments have demonstrated a 30% reduction in analysis time compared to traditional methods. This work highlights the potential of automation, adaptability, and decentralized architectures in modern DF a step towards distributed digital forensics.

Social networking platforms like X (Twitter) serve as hubs for open human interaction, but they are also increasingly infiltrated by automated accounts masquerading as human users. These bots often engage in activities such as spreading fake news and manipulating public opinion during politically sensitive times like elections. Most of the current bot detection methods rely on black-box algorithms, raising concerns about their transparency and practical usability. This study aims to address these limitations by developing a novel methodology for the detection of spambots and fake followers using annotated data. To this end, we propose an interpretable machine learning (ML) framework, leveraging multiple ML algorithms with hyperparameters optimized through cross-validation, to enhance the detection process. Furthermore, we analyze several features and provide a unique feature set that is optimized to offer excellent performance for bot detection. Moreover, we utilize multiple interpretable AI techniques which include Shapley Additive Explanations (SHAP) and Local Interpretable Model-agnostic Explanations (LIME). SHAP will help to display the effects of particular characteristics on the model’s prediction which will help in determining whether an account is a bot or a legitimate user. LIME will help to comprehend the model’s predictions, offering clarity regarding the traits or attributes that drive the classification conclusion. LIME allows researchers to detect bot-like activity in social networks by generating locally faithful explanations for each prediction. Our model offers enhanced interpretability by clearly highlighting the impact of various features used for spam and fake follower detection when compared to existing state-of-the-art social network bot detection systems. The results showcase the model’s ability to identify key distinguishing attributes between bots and legitimate users which offers a transparent and effective solution for social network bot detection. Additionally, we utilize two comprehensive datasets including Cresci-15 and Cresci-17, which serve as robust baselines for comparison. Our model showcases its effectiveness by outperforming other methods while providing interpretability which increases performance and reliability for the task of bot detection. 

The continuous evolution of quantum computing has shown novel and transformative possibilities and critical implications for the Industrial Internet of Things (IIoT) forensic processes. With the potential to break traditional encryption algorithms and process diverse datasets at unprecedented speeds, quantum computing could disrupt current approaches to digital forensic evidence (DFE) collection, preservation, and hybrid quantum-classical data analysis methods across IIoT environments, an emerging topic in digital forensics. This paper proposes a generic quantum safe IIoT forensic (QS-IIoT-F) framework, explores the implications of quantum computing for IIoT forensics, mentions the opportunities of quantum computing in IIoT forensics, and future research directions. By addressing these issues, this paper aims to pave the way for future-proof IIoT forensic methodologies, ensuring the integrity, efficiency, and reliability of digital forensic investigations in IIoT in a quantum-powered era.This article is categorized under: Digital and Multimedia Science > Cyber Threat Intelligence Digital and Multimedia Science > IoT Forensics Digital and Multimedia Science > Cybercrime Investigation

In the Industrial Internet of Things (IIoT), the absence of standardized forensic processes presents substantial hurdles to effective investigations. As IIoT devices become ubiquitous in critical infrastructure, ensuring consistency and reliability in forensic procedures becomes imperative. This paper aims to front the critical need for standardized IIoT forensic processes, challenges, impact and industry and government roles in achieving this. The motivation for this study is underscored by recent incidents where the absence of standardization impeded forensic analysis, leading to delayed or inconclusive results. By proposing a taxonomy of forensic processes tailored to the IIoT landscape, this paper examines and reviews the challenges, impacts, and the roles of industry and government in achieving standardization. The proposed approach aims to significantly enhance the ability of investigators to conduct thorough forensic investigations, ultimately improving accountability, security, and resilience in IIoT ecosystems.

The exponential growth and sophistication of cybersecurity threats have shown a rather an uncontrollable approaches, thus rendering traditional static defenses increasingly inadequate. This has necessitated the need for adaptive and intelligent reinforcements. In response, the authors of this paper propose a conceptual Emergent Collective Cognitive Defense (E-CCD) network-a novel architectural framework that unifies meta-cognition, theory-of-mind (ToM), and Emergent Collective Intelligence (ECI) into a cohesive, adaptive defense system. Unlike conventional systems that react solely to known threats, the E-CCD network endows individual agents with the ability to self-assess, continuously learn from experience, and anticipate adversarial behavior through probabilistic reasoning. The primary outcomes of this work include; a detailed architectural framework herein referred as E-CCD network, mathematical formulations for its core components, and a description of how the E-CCD framework enables proactive cyber defense, possible applicability, and key implementation considerations. In addition, the roles of federated learning and consensus algorithms, reinforcement learning, adversarial training are explored too. It is the authors’ opinion that this research lays the groundwork for next-generation cybersecurity systems capable of adapting to evolving threats in real-time. 

Sketch-based image retrieval (SBIR) utilizes sketches to search for images containing similar objects or scenes. Due to the proliferation of touch-screen devices, sketching has become more accessible and therefore has received increasing attention. Deep learning has emerged as a potential tool for SBIR, allowing models to automatically extract image features and learn from large amounts of data. To the best of our knowledge, there is currently no systematic literature review (SLR) of SBIR with deep learning. Therefore, the aim of this review is to incorporate related works into a systematic study, highlighting the main contributions of individual researchers over the years, with a focus on past, present and future trends. To achieve the purpose of this study, 90 studies from 2016 to June 2023 in 4 databases were collected and analyzed using the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) framework. The specific models, datasets, evaluation metrics, and applications of deep learning in SBIR are discussed in detail. This study found that Convolutional Neural Networks (CNN) and Generative Adversarial Networks (GAN) are the most widely used deep learning methods for SBIR. A commonly used dataset is Sketchy, especially in the latest Zero-shot sketch-based image retrieval (ZS-SBIR) task. The results show that Mean Average Precision (mAP) is the most commonly used metric for quantitative evaluation of SBIR. Finally, we provide some future directions and guidance for researchers based on the results of this review. © 2013 IEEE.