Search Results (13)

Large language models (LLMs) have shown remarkable skills across various activities, including text generation and code synthesis. Their widespread applicability, however, raises substantial concerns about security, privacy, and possibly misuse. Of recent legislative efforts, the most notable is the proposed EU AI Act, which classifies specific AI applications as high-risk. For detailed regulatory guidance, also refer to the GDPR and HIPAA privacy rules. This paper introduces SecureLLM, a novel framework that integrates lightweight cryptographic protocols, decentralized fine-tuning strategies, and differential privacy to mitigate data leakage and adversarial attacks in LLM ecosystems. We propose SecureLLM as a conceptual security architecture for LLMs, offering a unified approach that can be adapted and tested in real-world deployments. While extensive empirical benchmarks are deferred to future studies, we include a small-scale demonstration illustrating how differential privacy can reduce membership inference risks with a manageable overhead. The SecureLLM framework underscores the potential of cryptography, differential privacy, and decentralized fine-tuning for creating safer and more compliant AI systems. Full article

Artificial intelligence (AI) has significantly driven advancement in the healthcare field by enabling the integration of highly advanced algorithms to improve diagnostics, patient surveillance, and treatment planning. Nonetheless, dependence on sensitive health data and automated decision-making exposes such systems to escalating risks of privacy breaches and is under rigorous regulatory oversight. In particular, the EU AI Act classifies AI uses pertaining to healthcare as “high-risk”, thus requiring the application of strict provisions related to transparency, safety, and privacy. This paper presents a comprehensive overview of the diverse privacy attacks that can target machine learning (ML)-based healthcare systems, including data-centric and model-centric attacks. We then propose a novel privacy-preserving architecture that integrates federated learning with secure computation protocols to minimally expose data while ensuring strong model performance. We outline an ongoing monitoring mechanism compliant with EU AI Act specifications and GDPR standards to further improve trust and compliance. We further elaborate on an independent adaptive algorithm that automatically tunes the level of cryptographic protection based on contextual factors like risk severity, computational capacity, and regulatory environment. This research aims to serve as a blueprint for designing trustworthy, high-risk AI systems in healthcare under emerging regulations by providing an in-depth review of ML-specific privacy threats and proposing a holistic technical solution. Full article

The increasing complexity of energy grids, driven by rising demand and unpredictable residential consumption, highlights the need for efficient demand response (DR) strategies and data-driven services. This paper proposes a machine learning-based framework for DR that clusters users based on their consumption patterns and categorizes individual usage into distinct profiles using K-means, Hierarchical Agglomerative Clustering, Spectral Clustering, and DBSCAN. Key features such as statistical, temporal, and behavioral characteristics are extracted, and the novel Household Daily Load (HDL) approach is used to identify residential consumption groups. The framework also includes context analysis to detect daily variations and peak usage periods for individual users. High-impact users, identified by anomalies such as frequent consumption spikes or grid instability risks using IsolationForest and kNN, are flagged. Additionally, a classification service integrates new users into the segmented portfolio. Experiments on real-world datasets demonstrate the framework’s effectiveness in helping energy managers design tailored DR programs. Full article

Integrating artificial intelligence into border control systems may help to strengthen security and make operations more efficient. For example, the emerging application of artificial intelligence for lie detection when inspecting passengers presents significant opportunities for future implementation. However, as it makes use of technology that is associated with artificial intelligence, the system is classified as high risk, in accordance with the EU AI Act and, therefore, must adhere to rigorous regulatory requirements to mitigate potential risks. This manuscript distinctly amalgamates the technical, ethical, and legal aspects, thereby offering an extensive examination of the AI-based lie detection systems utilized in border security. This academic paper is uniquely set apart from others because it undertakes a thorough investigation into the categorization of these emerging technologies in terms of the regulatory framework established by the EU AI Act, which classifies them as high risk. It further makes an assessment of practical case studies, including notable examples such as iBorderCtrl and AVATAR. This in-depth analysis seeks to emphasize not only the enormous challenges ahead for practitioners but also the progress made in this emerging field of study. Furthermore, it seeks to investigate threats, vulnerabilities, and privacy concerns associated with AI, while providing security controls to address difficulties related to lie detection. Finally, we propose a framework that encompasses the EU AI Act’s principles and serves as a foundation for future approaches and research projects. By analyzing current methodologies and considering future directions, the paper aims to provide a comprehensive understanding of the viability and consequences of deploying AI lie detection capabilities in border control. Full article

Breast cancer is the most common cancer in the world. With a 5-year survival rate of over 90% for patients at the early disease stages, the management of side-effects of breast cancer treatment has become a pressing issue. Observational, real-world data such as electronic health records, insurance claims, or data from wearable devices have the potential to support research on the quality of life (QoL) of breast cancer patients (BCPs), but care must be taken to avoid errors introduced due to data quality and bias. This paper proposes a causal inference methodology for using observational data to support research on the QoL of BCPs, focusing on the osteopenia of patients undergoing treatment with aromatase inhibitors (AIs). We propose a machine learning-based pipeline to estimate the average and conditional average treatment effects (ATE and CATE). For evaluation, we develop a Structural Causal Model for the osteopenia of BCPs and rely on synthetically generated data to study the effectiveness of the proposed methodology under various data challenges. A set of studies were designed to estimate the effect of high-intensity exercise on bone mineral density loss using synthetic datasets of BCPs under AI treatment. Four observational study scenarios were evaluated, corresponding to synthetically generated data of 1000 BCPs with (a) no bias, (b) sampling bias, (c) hidden confounder bias, and (d) bias due to unobserved mediator. In all cases, evaluations were performed under both complete and missing data scenarios. In particular, machine learning-based models based on tree ensembles and neural networks achieved a lower estimation error by 23.8–51.3% and 32.4–89.3% for ATE and CATE, respectively, compared to direct estimation using sample averages. The proposed approach shows improved effectiveness in treatment effect estimation in the presence of missing values and sampling bias, compared to a “traditional” statistical analysis workflow. This suggests that the application of causal effect estimation methods for the study of BCPs’ quality of life using real-world data is promising and worth pursuing further. Full article

Predicting errors in home appliances is crucial for maintaining the reliability and efficiency of smart homes. However, there is a significant lack of such data on appliance malfunctions that can be used in developing effective anomaly detection models. This research paper presents a novel approach for simulating errors of heterogeneous home appliance power consumption patterns. The proposed model takes normal consumption patterns as input and employs advanced algorithms to produce labeled anomalies, categorizing them based on the severity of malfunctions. One of the main objectives of this research involves developing models that can accurately reproduce anomaly power consumption patterns, highlighting anomalies related to major, minor, and specific malfunctions. The resulting dataset may serve as a valuable resource for training algorithms specifically tailored to detect and diagnose these errors in real-world scenarios. The outcomes of this research contribute significantly to the field of anomaly detection in smart home environments. The simulated datasets facilitate the development of predictive maintenance strategies, allowing for early detection and mitigation of appliance malfunctions. This proactive approach not only improves the reliability and lifespan of home appliances but also enhances energy efficiency, thereby reducing operational costs and environmental impact. Full article

Low-power embedded systems have been widely used in a variety of applications, allowing devices to efficiently collect and exchange data while minimizing energy consumption. However, the lack of extensive maintenance procedures designed specifically for low-power systems, coupled with constraints on anticipating faults and monitoring capacities, presents notable difficulties and intricacies in identifying failures and customized reaction mechanisms. The proposed approach seeks to address the gaps in current resource management frameworks and maintenance protocols for low-power embedded systems. Furthermore, this paper offers a trilateral framework that provides periodic prescriptions to stakeholders, a periodic control mechanism for automated actions and messages to prevent breakdowns, and a backup AI malfunction detection module to prevent the system from accessing any stress points. To evaluate the AI malfunction detection module approach, three novel autonomous embedded systems based on different ARM Cortex cores have been specifically designed and developed. Real-life results obtained from the testing of the proposed AI malfunction detection module in the developed embedded systems demonstrated outstanding performance, with metrics consistently exceeding 98%. This affirms the efficacy and reliability of the developed approach in enhancing the fault tolerance and maintenance capabilities of low-power embedded systems. Full article

The crucial need for perpetual monitoring of photovoltaic (PV) systems, particularly in remote areas where routine inspections are challenging, is of major importance. This paper introduces an advanced approach to optimizing the maximum power point while ensuring real-time PV error handling. The overarching problem of securing continuous monitoring of photovoltaic systems is highlighted, emphasizing the need for reliable performance, especially in remote and inaccessible locations. The proposed methodology employs an innovative genetic algorithm (GA) designed to optimize the maximum power point of photovoltaic systems. This approach takes into account critical PV parameters and constraints. The single-diode PV modeling process, based on environmental variables like outdoor temperature, illuminance, and irradiance, plays a pivotal role in the optimization process. To specifically address the challenge of perpetual monitoring, the paper introduces a technique for handling PV errors in real time using evolutionary-based optimization. The genetic algorithm is utilized to estimate the maximum power point, with the PV voltage and current calculated on the basis of simulated values. A meticulous comparison between the expected electrical output and the actual photovoltaic data is conducted to identify potential errors in the photovoltaic system. A user interface provides a dynamic display of the PV system’s real-time status, generating alerts when abnormal PV values are detected. Rigorous testing under real-world conditions, incorporating PV-monitored values and outdoor environmental parameters, demonstrates the remarkable accuracy of the genetic algorithm, surpassing 98% in predicting PV current, voltage, and power. This establishes the proposed algorithm as a potent solution for ensuring the perpetual and secure monitoring of PV systems, particularly in remote and challenging environments. Full article

The accurate assessment of visual comfort in indoor spaces is crucial for creating environments that enhance occupant well-being, productivity, and overall satisfaction. This paper presents a groundbreaking contribution to the field of visual comfort assessment in occupied buildings, addressing the existing research gap in methods for evaluating visual comfort once a building is in use while ensuring compliance with design specifications. The primary aim of this study was to introduce a pioneering approach for estimating visual comfort in indoor environments that is non-intrusive, practical, and can deliver accurate results without compromising accuracy. By incorporating mathematical visual comfort estimation into a regression model, the proposed method was evaluated and compared using real-life scenario. The experimental results demonstrated that the suggested model surpassed the mathematical model with an impressive performance improvement of 99%, requiring fewer computational resources and exhibiting a remarkable 95% faster processing time. Full article

Smart homes enhance energy efficiency without compromising residents’ comfort. To support smart home deployment and services, an IoT network must be established, while energy-management techniques must be applied to ensure energy efficiency. IoT networks must perpetually operate to ensure constant energy and indoor environmental monitoring. In this paper, an advanced sensor-agnostic plug-n-play prescriptive edge-to-edge IoT network management with micro-services is proposed, supporting also the semantic interoperability of multiple smart edge devices operating in the smart home network. Furthermore, IoT health-monitoring algorithms are applied to inspect network anomalies taking proper healing actions/prescriptions without the need to visit the residency. An autoencoder long short-term memory (AE-LSTM) is selected for detecting problematic situations, improving error prediction to 99.4%. Finally, indicative evaluation results reveal the mitigation of the IoT system breakdowns. Full article

In Fog Computing, fog colonies are formed by nodes cooperating to provide services to end-users. To enable efficient operation and seamless scalability of fog colonies, decentralized control over participating nodes should be promoted. In such cases, autonomous Fog Nodes operate independently, sharing the context in which all colony members provide their services. In the paper, we explore different techniques of context diffusion and knowledge sharing between autonomous Fog Nodes within a fog colony, using ECTORAS, a publish/subscribe protocol. With ECTORAS, nodes become actively aware of their operating context, share contextual information and exchange operational policies to achieve self-configuration, self-adaptation and context awareness in an intelligent manner. Two different ECTORAS implementations are studied, one offering centralized control with the existence of a message broker, to manage colony participants and available topics, and one fully decentralized, catering to the erratic topology that Fog Computing may produce. The two schemes are tested as the Fog Colony size is expanding in terms of performance and energy consumption, in a prototype implementation based on Raspberry Pi nodes for smart building management. Full article

Blockchains are attracting the attention of stakeholders in many industrial domains, including the logistics and supply chain industries. Blockchain technology can effectively contribute in recording every single asset throughout its flow on the supply chain, contribute in tracking orders, receipts, and payments, while track digital assets such as warranties and licenses in a unified and transparent way. The paper provides, through its methodology, a detailed analysis of the blockchain fit in the supply chain industry. It defines the specific elements of blockchain that affect supply chain such as scalability, performance, consensus mechanism, privacy considerations, location proof and cost, and details on the impact that blockchains will have in disrupting the supply chain industry. Discussing the tradeoff between consensus cost, throughput and validation time it proceeds with a suggested high-level architectural approach, and concludes as a result with a discussion on changes needed and challenges faced for an in-vivo deployment of blockchains in the supply chain industry. While the technological features of modern blockchains can effectively facilitate supply chain uses cases, the various challenges that still remain, bring in front of us a wide set of needed changes and further research efforts for achieving a global, production level blockchain for the supply chain industry. Full article