Data security and privacy are crucial for the Internet of Medical Things (IoMT) and the digitization of healthcare systems. IoMT offers a revolutionary approach to healthcare monitoring, enabling remote patient interaction, sensor data collection, and even in-body implants. However, its wireless nature creates significant security vulnerabilities, and robust security mechanisms are essential to ensure patient trust. Hence, a lightweight privacy preservation mechanism based on blockchain and consensus is developed with minimal computational overhead. Initially, the reputation score is computed for every registered node using the confidence threshold value and transaction to verify the trustworthiness of the IoMT nodes. Tiny Feistel Cipher Encryption technique (TFCEA), a lightweight cryptographic technique, has been proposed to encrypt IoMT data. Based on the computed trust score, an Adaptive Proof of Work (APoW) consensus algorithm has been used to regulate block creation. APoW implements the Genetically Modified Salp Swarm Optimization (GM-SSO) algorithm for the node selection. To make it lightweight, the experimentation used the lightweight cryptographic algorithm and fine-tuned consensus APOW to facilitate the addition of new blocks according to their trust score. In terms of performance analysis, the presented experiments are benchmarked with the well-established works in the candidate domain. Tested results reveal its economic affordability by attaining 243.513 ms for 500 records as an execution time. An informal security analysis has been carried out to justify the robustness of the presented framework against various attacks. To increase resilience against cyber-attacks and ensure reliability in healthcare data security, Modified Principal Component Analysis (MPCA) enabled anomaly detection models were built using the BoT-IoT dataset. Tested models achieve 99.99% and 98.75% accuracy, indicating that the developed model is more suitable to prevent the possible cyber-attacks anticipated in smart healthcare.
D. Kavitha, Kiruthika Venkataramani, N. R., S. Ravikumar
ABSTRACT The Internet of Things (IoT) is transforming numerous sectors but also presents unique security challenges due to its interconnected and resourceâconstrained devices. This study introduces the Bidirectional Gaussian Hummingbird Optimized EndâtoâEnd Blockchain (BGHOâE2EB) model, designed to detect and classify cyberattacks within IoT environments. Unlike preventive approaches, the developed model focuses on realâtime detection and categorization of attacks, enabling timely responses to emerging threats. The proposed model integrates blockchain technology through Ethereumâbased smart contracts to enhance the security and integrity of data exchanges within IoT networks. Additionally, a Gaussian Artificial Hummingbird Algorithm is employed for optimal feature selection, minimizing data dimensionality and computational load. A Bidirectional Long ShortâTerm Memory (BiâLSTM) network further improves the model's capability by accurately detecting and categorizing cyber threats based on selected features. The Adam optimizer is used for efficient parameter tuning within the BiâLSTM network, ensuring highâperformance cyberattack detection. The proposed model was evaluated using established IoT security benchmarks, including the UNSWâNB15, BOTâIoT, and NSLâKDD datasets, accomplishing an accuracy of 98.7%, precision of 96.3%, and security level of 99.5%, significantly outperforming traditional methods. These results demonstrate the effectiveness of BGHOâE2EB as a robust tool for detecting and classifying cyberattacks in IoT networks, making it suitable for realâworld deployment in dynamic IoT environments where security is paramount.
Internet of Things (IoT) devices become more and more important because they are useful in different applications, for example, traffic monitoring, public safety, and environmental management. However, the vastness and diversity of IoT data and the requirement for real-time decision-making create a big challenge for anomaly detection frameworks. In this work, we propose SwinIoT, a new framework with hierarchical and windowed attention mechanisms of Swin Transformer that is especially good for the purpose of behavioral anomaly detection in IoT settings. SwinIoT would solve important problems of class imbalance, noisy data, and heterogeneous devices through the introduction of custom attention models embedding real-time optimizations. The proposed framework was benchmarked on nine datasets such as ARAS, CASAS, WESAD, and UCF Crime compared to the above-mentioned state-of-the-art algorithms like Active Learning-Based Anomaly Detection, Deep Support Vector Data Description (DSVDD), Deep Support Vector Data Description Contractive Autoencoder (DSVDD-CAE), and Federated Principal Component Analysis (FedPCA). It is proven to be better than the above algorithms by attaining up to 96% accuracy, 97% Mean Average Precision$(mAP)$, and excellent Area Under the Receiver Operating Characteristic curve (AUC-ROC) as well as Precision-Recall (PR) metrics, especially in low-resource and unbalanced data scenarios. The results indicate the potential of SwinIoT in scalable and accurate anomaly detection to the development of safer, smarter cities, ensuring reliability and security in critical systems.
This masterâs thesis examines the use of large language models for zero-shot anomaly detection in alphanumeric vehicle datasets, filling a gap where traditional statistical methods face limitations. While numerical data can be reliably assessed with algorithms like Local Outlier Factor or Isolation Forest, the high-dimensional nature of alphanumeric serial numbers makes them difficult to model with established algorithms. Using an iterative design science approach, this study develops and tests a Proof-of-Concept Python application that uses state-of-the-art large language models to detect anomalies in real-world vehicle datasets. Besides some prompt engineering, the models are intentionally not fine-tuned, enabling application without in-depth knowledge of large language models. The theoretical background covers data management, anomaly detection, and the core principles of large language models. Results show that large language models, especially Googleâs Gemini 2.5 Pro, can effectively identify anomalies in both numerical and alphanumeric data. Compared to statistical algorithms, large language models offer the benefit of processing alphanumeric inputs, adding a valuable extension to the anomaly detection toolkit. However, challenges like hallucination, inconsistent length counting, and sensitivity to highly anomalous datasets highlight current limitations. Additionally, statistical methods remain more efficient, scalable, and cost-effective for purely numerical datasets. The findings confirm that large language models can be applied in a zero-shot manner to detect anomalies in alphanumeric datasets. Beyond the automotive industry, these insights can be applied to other fields where alphanumeric identifiers are essential. This work advances both academic discussion and practical applications, providing a foundation for future research on fine-tuned models and industrial implementation.
With the widespread deployment of Deep-Learning-as-a-Service, secure multi-party computation-based outsourcing neural network (NN) inference has garnered significant attention for its high-security guarantee. Nevertheless, under the dishonest-majority setting with malicious adversaries, prior secure inference works are still costly in terms of communication and run-time. Additionally, existing outsourcing frameworks impose a substantial client-side design, which leads to obstacles in resource-constrained devices. To address the above challenges, we propose MD-SONIC, an online efficient and maliciously-secure framework for outsourcing NN inference with a dishonest majority. We first construct communication-efficient n-party protocols for the basic primitives such as fixed-point multiplication and most significant bit extraction by combining mask-sharing and TinyOT-sharing with SPD$\mathbb {Z}_{2^{k}}$seamlessly. Then, we build fast secure blocks for the widely used NN operators, including matrix multiplication, ReLU, and Maxpool, on top of our basic primitives. To enable an arbitrary number of users to outsource the secure inference task to n computing servers, we propose a lightweight-client and fast$\Sigma $paradigm named SPIN, stemming from zero-knowledge proofs. Our SPIN can be instantiated into a set of efficient outsourcing protocols over multiple algebraic structures (e.g., finite field and ring). We also conduct extensive evaluations of MD-SONIC on various neural networks. Compared to the work by DamgĂ„rd et al. (IEEE S&Pâ19) and MD-ML (USENIX Securityâ24), we achieve up to$594.4\times $and$45.1\times $online communication improvements, and improve the online execution time by at most$14.3\times $(resp.$20.5\times $) and$1.8\times $(resp.$2.3\times $) in LAN (resp. WAN).
Mantri Christ Elison, Martin Victor K, Gifton Paul Immanuel
The objective of this research is to develop an R&D (Research and Development) for the hardiness relay alert system, including applying the machine learning, and the fuzzy logic networks for the real time Ethereum transaction 'match failure' detection and the improved Ethereum blockchain security.As an example, the system is computing on the transactions due to the fact the system for transaction analysis corresponds with concrete intrinsic characteristics and thus it mainly takes out suspicious or malicious transactions.The logistic regression, support vector machines (SVM) decision tree and random forests are used in this research and optimized by grid search.Finally, on the other hand, uncertainty problems and false alarms are solved where fuzzy membership functions are used to put transaction attributes into linguistic hobbled variables (such as 'low', 'medium' and 'high').The conclusion of this descriptive research is that fuzzy logic integration with machine learning can improve the approach of anomaly mediation compared to the rules based approach and it is superior to rules based approach.Finally, the effectiveness of the models is detailed and replicated in various graphical representations of the decision making process and membership functions to show that the system can be deployed in real time to secure blockchain networks.
Developers and users are drawn to Ethereum due to its rapidly growing asset count. However, potential vulnerabilities and malicious behaviors during the execution of smart contracts have led to an increasing demand for security detection technology. Conventional static and dynamic analysis methods are less useful in the case of complex opcode sequences and multiple execution paths. To tackle this problem, this paper proposes an Ethereum intrusion detection method based on Bidirectional Long Short-Term Memory (Bi-LSTM) network with multi-head attention. It examines the opcode execution paths generated from the intra-and-inter-function Control Flow Graphs (CFGs) using the EPP algorithm and captures the rich feature representations and long dependencies. This combination increases the precision and efficacy of detecting malicious activity and smart contract vulnerabilities while simultaneously enhancing the modelâs robustness and interpretability and handling variable-length sequences. For the five selected vulnerabilities, the precision, recall and F1-score of this model are above 89.9%, 87.3%, and 88%, respectively.
Hanbiao Du, Meng Shen, Yang Liu, Zheng Che · 7 authors
Ethereum serves as the cornerstone for value transfer in Web 3.0, providing a decentralized and efficient trust mechanism for global connectivity. However, the anonymity of Ethereum undermines market regulatory capabilities, leading to frequent malicious behaviors such as Ponzi Scheme, Money Laundering, and Phishing. Therefore, in the face of the diverse and continuously emerging malicious behaviors, implementing fine-grained detection is crucial for maintaining the prosperous development of the blockchain ecosystem. In this paper, we propose FiMAD, a fine-grained and class-incremental malicious account detection framework based on dynamic graph learning. Specifically, we first propose a general graph structure calledDynamic Account Relation Graph (DARG), which dynamically models Ethereum accounts from a continuous-time perspective. Then, we design a cascade graph feature extraction method to capture deep temporal evolution patterns and neighbor interaction features in DARG. Next, we construct a pre-training universal encoder to transform account features into high-dimensional embeddings, followed by fine-tuning the model classifier with a few labeled samples, enabling accurate fine-grained detection and rapid updates for incremental classes. We conduct extensive experiments using real Ethereum data. The results demonstrate that FiMAD outperforms state-of-the-art (SOTA) methods in fine-grained detection across five typical scenarios: class-incremental, full data, new malicious accounts, imbalanced data, and binary classification. In the class-incremental scenario, FiMAD improves the Macro-F1 by up to 26.4% compared to SOTA methods.
Bitcoin is the most valuable cryptocurrency and is renowned for its rapid and volatile price fluctuations in comparison to other currencies. This offers potential for the prediction of Bitcoin prices and has attracted the interest of researchers. Twitter (X) is one of the most widely used social media platforms. The aim of this study is to analyse the sentiment expressed in comments about bitcoin on the social media platform X using a variety of machine learning algorithms. A variety of machine learning techniques are used to classify user sentiment towards bitcoin. Moreover, the efficacy of standard bag-of-words and term frequency-inverse document frequency (TF-IDF) methods is evaluated in comparison with machine learning approaches for the purpose of expressing text as numerical vectors. Finally, a keyword ranking was performed to determine the importance of each sentiment in the development of cryptocurrencies. The bag-of-words and TF-IDF methods were used, which facilitate the representation of text-based data. The best result was obtained with the decision trees algorithm (98.74% accuracy) using the TF-IDF method. The bag-of-words method was found to produce better results in general.
Mohamed A. Fouly, Taysir Hassan A. Soliman, Ahmed I. Taloba
A blockchain is made up of an ordered list of nodes connected by links known as chains. The nodes in the blockchain store data and are stored together. The distributed and decentralized ledger technology, blockchain, empowers cryptocurrencies like Bitcoin and Ethereum. It makes a safe and open record of transactions by permitting the distribution of digital data as a âblockâ but prohibiting its duplication. Furthermore, blockchain-based anomaly detection tools, which always automatically detect and weed out abnormal behaviors, are essential for protecting networks and systems from unforeseen intrusions. A smart contract could monitor real-time transaction volumes, access patterns, or resource usage. If anomalies are detected, such as unusual spikes in activity, the smart contract can trigger alerts or take predefined actions. Numerous anomaly detection analysis techniques have been put out and used in the scientific literature in various fields. This paper provides an overview of the latest machine learning techniques for identifying abnormal behaviors in blockchain, such as supervised, unsupervised, and deep learning. We also go over a few of the applications for anomaly behaviors detection.
This study investigates the critical challenges associated with ensuring the security and robustness of artificial intelligence (AI) systems, especially within high-stakes applications such as autonomous vehicles, healthcare, and financial technologies. The primary objective is to identify vulnerabilities in AI algorithms and propose effective mitigation strategies. The research emphasizes contemporary threats, including adversarial attacks, algorithmic opacity, data breaches, and the ethical ramifications of AI deployment. A review of current literature reveals that adversarial attacks, where subtle input perturbations cause significant misclassifications, present a considerable risk to AI reliability. Techniques such as robust training, involving training models on adversarial examples, have shown effectiveness in improving resilience, albeit with higher computational demands. The study also explores the importance of explainable AI (XAI) tools like LIME and SHAP, which enhance transparency by clarifying the decision-making processes of complex models. This transparency is vital for fostering user trust, especially in fields like medicine and finance, where understanding AI decisions is essential. XAI approaches enable better oversight and adherence to ethical standards. Data privacy concerns are addressed through methods such as differential privacy, which protects sensitive information by adding noise, and federated learning, which enables decentralized model training without exposing raw data. The findings indicate that these strategies secure data while maintaining model efficacy. By integrating robustness and explainability, this study contributes practical solutions to strengthen AI systems against evolving threats, advancing AI security and fostering trust in these technologies.
The rapid evolution of cyber threats, driven by artificial intelligence (AI) and machine learning (ML), has exposed critical gaps in traditional cybersecurity frameworks, particularly in real-time threat detection and response. This paper presents research on the Next-Gen Cyber Security Sentinel, a system that integrates AI algorithms with blockchain-based smart contracts to detect and mitigate sophisticated, AI-powered cyberattacks in real-time. Through the use of Software-Defined Networks (SDN), the system simulates complex attack scenarios, providing flexible and scalable network management. Penetration testing confirmed the system's ability to detect, respond to, and mitigate advanced cyber threats, ensuring enhanced protection of critical infrastructure. The findings were both novel and innovative, demonstrating that the integration of AI and blockchain technology significantly improves the speed and accuracy of real-time threat detection. This research contributes to the field of cybersecurity by offering a robust, scalable solution to counter emerging AI-driven attacks, particularly in critical sectors such as smart cities and Industry 4.0 environments. These findings offer crucial insights for advancing cybersecurity solutions in the face of rapidly evolving, AI-driven cyber threats. The AI-Blockchain platform achieved a 95% detection rate with a 2% false positive rate and maintained blockchain transaction latency under 200 milliseconds, demonstrating significant improvements in real-time threat detection and response capabilities.
The research introduces a fresh hybrid architecture integrating Artificial Intelligence (AI) and Blockchain technology to achieve real-time anomaly detection and to ensure data integrity in medical scenarios. The LSTM-CNN motivated model returned results of 95% accuracy and 94% F1-score along with 93% recall, underscoring its enhanced capacity for spotting both illegal data access and suspicious medical dealings. The experimental setup led to an analysis of a synthetic hospital dataset comprising 26,000 items, which included patient admissions, billing transactions, and medical records, with 5.3 % of the data specifically designed to include anomalies to assess the model's performance. After deploying the Blockchain on Ethereum, we ensured that the data was immutable and secure, completely removing cases of data tampering and unauthorized alterations, which descended from 5 and 8 incidents before the installation to zero following it. Also, the solution showed that reductions in validation failures went from 7 to 1, illuminating how smart contracts automate data validation processes. Results from latency studies demonstrate that the system is capable of managing real-time transactions with an average delay of 3.2 seconds, establishing its suitability for fluid medical environments that demand fast access to data. The research points out that the proposed framework may considerably increase data security and operational efficiency within healthcare contexts. The study generates a firm framework for developing more secure and trustworthy hospital data management systems, while future efforts will concentrate on enhancing the framework for broader deployment and improving its scalability to deal with increasingly complex healthcare data scenarios.
Shabnam Fazliani, Mohammad Mowlavi Sorond, Arsalan Masoudifard
The advent of smart contracts has enabled the rapid rise of Decentralized Finance (DeFi) on the Ethereum blockchain, offering substantial rewards in financial innovation and inclusivity. This growth, however, is accompanied by significant security risks such as illicit accounts engaged in fraud. Effective detection is further limited by the scarcity of labeled data and the evolving tactics of malicious accounts. To address these challenges with a robust solution for safeguarding the DeFi ecosystem, we propose $\textbf{SLEID}$, a $\textbf{S}$elf-$\textbf{L}$earning $\textbf{E}$nsemble-based $\textbf{I}$llicit account $\textbf{D}$etection framework. SLEID uses an Isolation Forest model for initial outlier detection and a self-training mechanism to iteratively generate pseudo-labels for unlabeled accounts, enhancing detection accuracy. Experiments on 6,903,860 Ethereum transactions with extensive DeFi interaction coverage demonstrate that SLEID significantly outperforms supervised and semi-supervised baselines with $\textbf{+2.56}$ percentage-point precision, comparable recall, and $\textbf{+0.90}$ percentage-point F1 -- particularly for the minority illicit class -- alongside $\textbf{+3.74}$ percentage-points higher accuracy and improvements in PR-AUC, while substantially reducing reliance on labeled data.