Maritime port operations face escalating fraud risks, specifically cargo theft and document falsification, driven by fragmented paper-based processes and information silos. Conventional detection mechanisms lack real-time authentication for high-volume throughput. This study proposes an NFC-Blockchain integration framework that combines Near Field Communication with blockchain to create a digital twin for physical assets. Moving beyond descriptive analyses, this paper details a functional prototype from NFC tag initialization to smart contract execution on a Layer 1 blockchain, establishing a concrete system workflow for secure asset tracking and document verification. A mixed-methods design was utilized, combining a quantitative survey of 423 port operators with five expert interviews and case study analysis. Results reveal a significant readiness gap: while blockchain awareness is high (70%), NFC familiarity is low (35%), with adoption constrained by high implementation costs and technical expertise deficits (both 55%). The proposed architecture demonstrates that integrating these technologies automates compliance via smart contracts and provides tamper-proof real-time traceability. These findings offer actionable technical and strategic guidance for port authorities to mitigate fraud through decentralized infrastructure.
The exponential growth of the Internet of Things (IoT) has led to new cybersecurity threats from the rise in the number of various and resource-limited connected devices. Conventional intrusion detection systems (IDSs) may not be efficient in dealing with real-time, high-dimensional, class-imbalanced nonstationary IoT traffic data. To overcome these limitations, in this paper, an IoT intrusion detection framework is proposed based on Mixed Dung Beetle Optimisation with Deep Learning Cyber Security (MDBODR-DLCS). To this end, Z-score normalisation is first performed to normalise the features of network traffic, which enhances the stability of subsequent models. Then, Mixed Dung Beetle Optimisation (MDBO) is used for optimal feature selection and dimensionality reduction to remove redundant and irrelevant attributes. To employ the established temporal dependency information and contextual traffic patterns as features for intrusion classification, our proposed method utilises an Attention-Based Bidirectional Recurrent Neural Network (ABiRNN). Besides, Artificial Rabbits Optimisation (ARO) is employed to fine-tune hyperparameters to improve classification accuracy and convergence speed. We conduct an experimental evaluation on the WSNDS dataset that shows that our framework is able to outperform state-of-the-art intrusion detection solutions, with over 99% accuracy and very high sensitivity, specificity, F-score, and AUC for multiple attack classes. The comparative analyses showed that the proposed method outperforms other existing methods using machine learning and deep learning-based IDS structures while maintaining low computational complexity with an average execution time of 0.60 seconds per instance. The results were found to demonstrate that the developed method MDBODR-DLCS efficiently achieves high accuracy performance in real time and scalability, while being resilient due to its ability to detect even the variations an attacker might introduce, whilst maintaining very low overhead, making it appropriate for resource-constrained environments such as IoT.
Multi-label text classification remains challenging due to class imbalance, overlapping label distributions, and inter-label dependencies. Existing loss functions often address these issues independently, resulting in limited robustness and inconsistent generalization. To address these limitations, this paper proposes ConFoBCE-MLTC, a correlation-aware composite loss function that integrates binary cross-entropy, focal modulation, contrastive learning, and label correlation alignment within a unified optimization framework. The proposed method improves minority-class sensitivity, enhances feature separability, and explicitly models label co-occurrence structures without requiring architectural modifications. The approach was evaluated using multiple transformer-based architectures, including BERT, RoBERTa, DistilBERT, and mBERT, across both high-resource and low-resource datasets, including the newly constructed Hausa Emotion Corpus (HaEmoC_V1). Experimental results demonstrate consistent improvements over conventional loss functions and recent asymmetric loss formulations. In particular, mBERT combined with ConFoBCE-MLTC achieved the strongest overall performance, reaching 82.24 Micro-F1, 80.18 Macro-F1, and 73.31 Jaccard Score on the newly constructed dataset (HaEmoC_V1). Additional ablation and sensitivity analyses confirm the complementary contribution of focal weighting, contrastive representation learning, and label correlation alignment. These findings demonstrate the effectiveness of the proposed framework for multilingual multi-label emotion classification, particularly in low-resource settings.
The shift to biometric systems where unique qualities of a user used in place of standard passwords has brought a significant change in cybersecurity. The biometrics which consist of patterns of finger knuckles and fingernails give accurate and reliable identity recognition, and therefore minimize chances of unauthorized access and impersonation. The massive advancement of deep learning models allows the extraction of accurate biometric information in the images with the help of convolutional neural networks (CNNs), which are after that integrated to highly discriminative and consistent feature space. This high level of integration can help to build greater recognition strength and the safety of cryptographic keys that are produced by biometric features which is why this strategy can be chosen as one of the cornerstones in developing the safe biometric recognition systems that can help to build cybersecurity in the contemporary settings. This research introduces an enhanced cybersecurity for a dorsal finger knuckle pattern and finger nail recognition system, namely (ECS-DFKPFNRS), depending on an integrated feature deep learning (DL) with the cryptography algorithm. This ECS-DFKPFNRS utilizes nineteen key components of the dorsal finger knuckle pattern of ten fingers for concatenated fusion recognition of all the features extracted. This system applies a concatenated DL model to four pre-trained models: InceptionNetV3, ResNet50, DenseNet201, and MobileNet-V3. To conduct this, the ECS-DFKPFNRS utilizes a new segmentation method that relies on the Hands Landmark Module (MediaPipe Module) to detect key knuckle hand components (Five-Basic, Five-Major, Four-Minor, and the fingernails) for both hands. After that, the concatenated fusion feature vectors extracted from the concatenated DL model will be utilized to generate a symmetric key (secret key) by the Rubik's cube technique. This technique will be improved by incorporating logistic scrambling, permutations, and SHA-512 hashing; subsequently, these keys to generated after the enhanced Rubik’s cube technique and will be utilized to encrypt biometric data by employing the lightweight Grain-128a algorithm. This is an integrated deep feature for FKP and FN and encryption, aiming to guarantee confidentiality and integrity against statistical attacks. 11,076K Hands and the ’Poly U HD’ datasets were utilized to analyze and evaluate this system. The system which was proposed was more effective on these datasets with outstanding security measures: NPCR (99.95% and 99.63%), UACI (30.23% and 30.21%), and resistance to statistical attacks. It is demonstrated in this study that the combination of DL and biometric encryption improves the reliability of biometric data and biometric robustness, which improves cybersecurity and digital biometric identity protection.
Maritime port operations face escalating fraud risks, specifically cargo theft and document falsification, driven by fragmented paper-based processes and information silos. Conventional detection mechanisms lack real-time authentication for high-volume throughput. This study proposes an NFC-Blockchain integration framework that combines Near Field Communication with blockchain to create a digital twin for physical assets. Moving beyond descriptive analyses, this paper details a functional prototype from NFC tag initialization to smart contract execution on a Layer 1 blockchain, establishing a concrete system workflow for secure asset tracking and document verification. A mixed-methods design was utilized, combining a quantitative survey of 423 port operators with five expert interviews and case study analysis. Results reveal a significant readiness gap: while blockchain awareness is high (70%), NFC familiarity is low (35%), with adoption constrained by high implementation costs and technical expertise deficits (both 55%). The proposed architecture demonstrates that integrating these technologies automates compliance via smart contracts and provides tamper-proof real-time traceability. These findings offer actionable technical and strategic guidance for port authorities to mitigate fraud through decentralized infrastructure.
