Issue Of Trustworthiness Research Articles

Physics and Data Collaborative Root Cause Analysis

Data-driven tools for asset health management face significant challenges, including a lack of understanding of physical principles, difficulty incorporating domain experts’ experiences, and consequently low detection accuracy, leading to trustworthiness issues. Automatically integrating data-driven analysis with human knowledge and experience, as found in literature and maintenance logs, is critically needed. Recent progress in large language models (LLMs) offers opportunities to achieve this goal. However, there is still a lack of work that effectively combines pretrained LLMs with data-driven models for asset health management using industrial time series data as input. This paper presents a framework that integrates our recently proposed data-driven AI with pretrained LLMs to address root cause detection in industrial failure analysis. The framework employs LLMs to analyze outputs from our data-driven root cause analysis models, filtering out less relevant results and prioritizing those that align closely with physical principles and domain expertise. Our innovative approach leverages advanced data-driven analytics and a multi-LLM debate for collaborative decision-making, seamlessly merging data-driven insights with domain knowledge. Specifically, through our proposed self-exclusionary debates among multiple LLMs, biases inherent in single-LLM systems are effectively mitigated, enhancing reliability and stability. Crucially, the framework bridges the gap between data-driven models and physics-informed LLMs, accelerating the interaction between data and knowledge for more informed and realistic decision-making processes.

Adversarial Examples on XAI-Enabled DT for Smart Healthcare Systems.

There have recently been rapid developments in smart healthcare systems, such as precision diagnosis, smart diet management, and drug discovery. These systems require the integration of the Internet of Things (IoT) for data acquisition, Digital Twins (DT) for data representation into a digital replica and Artificial Intelligence (AI) for decision-making. DT is a digital copy or replica of physical entities (e.g., patients), one of the emerging technologies that enable the advancement of smart healthcare systems. AI and Machine Learning (ML) offer great benefits to DT-based smart healthcare systems. They also pose certain risks, including security risks, and bring up issues of fairness, trustworthiness, explainability, and interpretability. One of the challenges that still make the full adaptation of AI/ML in healthcare questionable is the explainability of AI (XAI) and interpretability of ML (IML). Although the study of the explainability and interpretability of AI/ML is now a trend, there is a lack of research on the security of XAI-enabled DT for smart healthcare systems. Existing studies limit their focus to either the security of XAI or DT. This paper provides a brief overview of the research on the security of XAI-enabled DT for smart healthcare systems. It also explores potential adversarial attacks against XAI-enabled DT for smart healthcare systems. Additionally, it proposes a framework for designing XAI-enabled DT for smart healthcare systems that are secure and trusted.

Network and cybersecurity applications of defense in adversarial attacks: A state-of-the-art using machine learning and deep learning methods

Abstract This study aims to perform a thorough systematic review investigating and synthesizing existing research on defense strategies and methodologies in adversarial attacks using machine learning (ML) and deep learning methods. A methodology was conducted to guarantee a thorough literature analysis of the studies using sources such as ScienceDirect, Scopus, IEEE Xplore, and Web of Science. A question was shaped to retrieve articles published from 2019 to April 2024, which ultimately produced a total of 704 papers. A rigorous screening, deduplication, and matching of the inclusion and exclusion criteria were followed, and hence 42 studies were included in the quantitative synthesis. The considered papers were categorized into a coherent and systematic classification including three categories: security enhancement techniques, adversarial attack strategies and defense mechanisms, and innovative security mechanisms and solutions. In this article, we have presented a systematic and comprehensive analysis of earlier studies and opened the door to potential future studies by discussing in depth four challenges and motivations of adversarial attacks, while three recommendations have been discussed. A systematic science mapping analysis was also performed to reorganize and summarize the results of studies to address the issues of trustworthiness. Moreover, this research covers a large variety of network and cybersecurity applications of defense in adversarial attack subjects, including intrusion detection systems, anomaly detection, ML-based defenses, and cryptographic techniques. The relevant conclusions well demonstrate what have achieved in defense mechanisms against adversarial attacks. In addition, the analysis revealed a few emerging tendencies and deficiencies in the area to be remedied through better and more dependable mitigation methods against advanced persistent threats. The findings of this review have crucial implications for the community of researchers, practitioners, and policy makers in network and cybersecurity using artificial intelligence applications.

A survey of safety and trustworthiness of large language models through the lens of verification and validation

Large language models (LLMs) have exploded a new heatwave of AI for their ability to engage end-users in human-level conversations with detailed and articulate answers across many knowledge domains. In response to their fast adoption in many industrial applications, this survey concerns their safety and trustworthiness. First, we review known vulnerabilities and limitations of the LLMs, categorising them into inherent issues, attacks, and unintended bugs. Then, we consider if and how the Verification and Validation (V&V) techniques, which have been widely developed for traditional software and deep learning models such as convolutional neural networks as independent processes to check the alignment of their implementations against the specifications, can be integrated and further extended throughout the lifecycle of the LLMs to provide rigorous analysis to the safety and trustworthiness of LLMs and their applications. Specifically, we consider four complementary techniques: falsification and evaluation, verification, runtime monitoring, and regulations and ethical use. In total, 370+ references are considered to support the quick understanding of the safety and trustworthiness issues from the perspective of V&V. While intensive research has been conducted to identify the safety and trustworthiness issues, rigorous yet practical methods are called for to ensure the alignment of LLMs with safety and trustworthiness requirements.

Distributed Group Key Management Based on Blockchain

Against the backdrop of rapidly advancing cloud storage technology, as well as 5G and 6G communication technologies, group key management faces increasingly daunting challenges. Traditional key management encounters difficulties in key distribution, security threats, management complexity, and issues of trustworthiness. Particularly in scenarios with a large number of members or frequent member turnover within groups, this may lead to security vulnerabilities such as permission confusion, exacerbating the security risks and management complexity faced by the system. To address these issues, this paper utilizes blockchain technology to achieve distributed storage and management of group keys. This solution combines key management with the distributed characteristics of blockchain, enhancing scalability, and enabling tracking of malicious members. Simultaneously, by integrating intelligent authentication mechanisms and lightweight data update mechanisms, it effectively enhances the security, trustworthiness, and scalability of the key management system. This provides important technical support for constructing a more secure and reliable network environment.

But was it Trustworthy? Methodological Experiences From a Study of a Hard-to-Reach Group of Students in Need of a Flexible Research Approach

Pedagogical research on experiences of learning among students with severe speech and physical impairment (SSPI) is sparse. This may be due to a lack of research-on-research methodology literature about students with SSPI, as they are difficult to find and there are barriers to their participation in mainstream research. Hence, method development is especially important regarding these students, who cannot participate when traditional inquiry methods are used. This article’s objective is therefore to advance method development by means of a retrospective investigation. The empirical findings consist of documented experiences from a previous study of students with SSPI (henceforth, “literacy study”). A computer-assisted email dialogue technique was developed in the literacy study’s pilot study and eventually used in the main study, to investigate the students’ experiences of their literacy development. The aim of this study is to retrospectively and critically examine the scientific trustworthiness of a methodological research approach based on an email dialogue technique used exploratively in the literacy study, to investigate the literacy development among the students grounded in their own experiences, and to contribute methodological experiences gained from that study regarding the relationship between the use of verification strategies and checking techniques. The computer-assisted email dialogue approach was necessary because the few participants were spread over great geographical distances. The approach was developed as an explorative and flexible inclusive research design and was used within the tradition of participatory research. The students in both the pilot and main studies (8–16 years of age) were treated as collaborators rather than research subjects. Both the verification strategies and techniques regarding trustworthiness criteria were found to be important for trustworthiness. The main conclusion, based on our experiences in this retrospective investigation, is that it is necessary to continuously and thoroughly focus on trustworthiness issues throughout the research process to obtain trustworthy findings.

Sampling and Trustworthiness Issues in Qualitative Research

Qualitative research is crucial in exploring the complexities of human experiences, behaviors, perceptions, and social phenomena. It is particularly effective in generating hypotheses, exploring new research topics, and capturing the subjective aspects of human interaction and experience. It emphasizes social, economic, and political context, cultural nuances, and participants' voices for comprehensive and holistic understanding. Determining an appropriate sampling method and adequacy of sample size remains a challenging aspect of qualitative research methodology. This paper highlights the key issues related to sampling approaches, sample size, and trustworthiness in qualitative research.

Faire: Repairing Fairness of Neural Networks via Neuron Condition Synthesis

Deep Neural Networks (DNNs) have achieved tremendous success in many applications, while it has been demonstrated that DNNs can exhibit some undesirable behaviors on concerns such as robustness, privacy, and other trustworthiness issues. Among them, fairness (i.e., non-discrimination) is one important property, especially when they are applied to some sensitive applications (e.g., finance and employment). However, DNNs easily learn spurious correlations between protected attributes (e.g., age, gender, race) and the classification task and develop discriminatory behaviors if the training data is imbalanced. Such discriminatory decisions in sensitive applications would introduce severe social impacts. To expose potential discrimination problems in DNNs before putting them in use, some testing techniques have been proposed to identify the discriminatory instances (i.e., instances that show defined discrimination 1 ). However, how to repair DNNs after detecting such discrimination is still challenging. Existing techniques mainly rely on retraining on a large number of discriminatory instances generated by testing methods, which requires huge time overhead and makes the repairing inefficient. In this work, we propose the method Faire to effectively and efficiently repair the fairness issues of DNNs, without using additional data (e.g., discriminatory instances). Our basic idea is inspired by the traditional program repair method that synthesizes proper condition checking. To repair traditional programs, a typical method is to localize the program defects and repair the program logic by adding condition checking. Similarly, for DNNs, we try to understand the unfair logic and reformulate it with well-designed condition checking. In this article, we synthesize the condition that can reduce the effect of features relevant to the protected attributes in the DNN. Specifically, we first perform the neuron-based analysis and check the functionalities of neurons to identify neurons whose outputs could be regarded as features relevant to protected attributes and original tasks. Then a new condition layer is added after each hidden layer to penalize neurons that are accountable for the protected features (i.e., intermediate features relevant to protected attributes) and promote neurons that are accountable for the non-protected features (i.e., intermediate features relevant to original tasks). In sum, the repair rate 2 of Faire reaches up to more than 99%, which outperforms other methods, and the whole repairing process only takes no more than 340 s. The evaluation results demonstrate that our approach can effectively and efficiently repair the individual discriminatory instances of the target model.

A Discussion of Key Aspects and Trends in Self Driving Vehicle Technology

Autonomous vehicles use remote-sensing technologies such as radar, GPS, cameras, and lidar to effectively observe their immediate environment and construct a comprehensive three-dimensional representation. The conventional constituents of this particular environment include structures, additional vehicles, people, as well as signage and traffic indicators. At now, a self-driving car is equipped with a wide array of sensors that are not found in a traditional automobile. Commonly used sensors include lasers and visual sensors, which serve the purpose of acquiring comprehensive understanding of the immediate environment. The cost of these sensors is high and they exhibit selectivity in their use requirements. The installation of these sensors in a mobile vehicle also significantly diminishes their operational longevity. Furthermore, the issue of trustworthiness is a matter of significant concern. The present article is structured into distinct parts, each of which delves into a significant aspect and obstacle pertaining to the trend and development of autonomous vehicles. The parts describing the obstacles in the development of autonomous vehicles define the conflict arising from the use of cameras and LiDAR technology, the influence of social norms, the impact of human psychology, and the legal complexities involved.

Reliability Analysis of the Functional Capabilities of an Autonomous Vehicle

The reliability of autonomous vehicles (AVs) is a research domain of high interest, covering a diverse pool of researchers, captains of smart auto industries, government agencies, and technology enthusiasts. The reliability of AVs is not extensively explored in the literature, despite the apprehension due to fatal accidents recorded in the past. Despite being in existence for over a decade, AVs have yet to reach a certified commercial-level deployment. Due to the complexity that comes with the self-operation of an AV, the issue of trustworthiness, which signifies reliability, becomes inevitable. The identification, analysis, and categorization of functional elements using systems engineering conceptual design principles and the linkage of these to the road traffic rules were conducted to address this. Also, the evaluation of the reliability of AVs using various developed vehicles from selected industries was addressed by integrating the traffic rules. The analysis of reliability was carried out using life-to-failure data premised on the probability plotting approach. It was found that there is a 99.94% chance that an autonomous vehicle will fail at least one of the traffic rules within 20 minutes of driving. Furthermore, the hazard rate of AVs was found to be on the rise, meaning a high indication of accidents.

POTENTIAL OPPORTUNITIES OF BLOCKCHAIN TECHNOLOGY IN STREAMLINING THE MALAYSIAN COMMERCIAL OFFICE BUILDING OPERATION MANAGEMENT PROCESS

Managing commercial real estate is extremely complex and cumbersome and involves many stakeholders who provide access to and modify a variety of information. This led to some common issues with traditional commercial real estate management such as lack of transparency, third-party interference, the complexity of the agreements, record keeping and security, and absence of real-time data. Accordingly, blockchain technology is anticipated to address the issues of data openness, trustworthiness, and correctness in the global real estate market. This research aims to discover the potential opportunities that blockchain could offer in streamlining the commercial office building operation management process. A conceptual framework of blockchain-based technology adoption in the current Malaysian commercial office building operation management process was developed based on the literature review and focus group discussion (FGD). The framework was underpinned by four core operation management that emerge from the research findings including financial, administrative, technical, and promotional management.

A transformative solution for construction safety: Blockchain-based system for accident information management

Effective management of accident information is a crucial component of safety management within the construction industry, as it reflects the safety performance of the company and allows them to identify the root causes of accidents and prevent similar accidents in the future. However, existing safety information systems provide self-owned, isolated, and centralized environments and fail to present a secure, transparent, and trustworthy platform for monitoring and management of accident information. To address these issues, this paper presents a novel decentralized blockchain-based system for accident/incident information management of construction projects. The proposed system leverages the benefits and advantages of blockchain, smart contracts, and decentralized IPFS storage to address the security transparency, tampering, and trustworthiness issues of the conventional approaches. The proposed system is simulated by using real-world construction accident data to demonstrate how blockchain technology can provide a novel solution to assure security, transparency, authenticity, availability, and immutability of the accident/incident data for improving safety management.

Trustworthy Medical Image Segmentation with improved performance for in-distribution samples

Despite the enormous achievements of Deep Learning (DL) based models, their non-transparent nature led to restricted applicability and distrusted predictions. Such predictions emerge from erroneous In-Distribution (ID) and Out-Of-Distribution (OOD) samples, which results in disastrous effects in the medical domain, specifically in Medical Image Segmentation (MIS). To mitigate such effects, several existing works accomplish OOD sample detection; however, the trustworthiness issues from ID samples still require thorough investigation. To this end, a novel method TrustMIS (Trustworthy Medical Image Segmentation) is proposed in this paper, which provides the trustworthiness and improved performance of ID samples for DL-based MIS models. TrustMIS works in three folds: IT (Investigating Trustworthiness), INT (Improving Non-Trustworthy prediction) and CSO (Classifier Switching Operation). Initially, the IT method investigates the trustworthiness of MIS by leveraging similar characteristics and consistency analysis of input and its variants. Subsequently, the INT method employs the IT method to improve the performance of the MIS model. It leverages the observation that an input providing erroneous segmentation can provide correct segmentation with rotated input. Eventually, the CSO method employs the INT method to scrutinise several MIS models and selects the model that delivers the most trustworthy prediction. The experiments conducted on publicly available datasets using well-known MIS models reveal that TrustMIS has successfully provided a trustworthiness measure, outperformed the existing methods, and improved the performance of state-of-the-art MIS models. Our implementation is available at https://github.com/SnehaShukla937/TrustMIS.

Contribution of Classroom Climate to the Teaching of Agriculture at Senior Secondary Schools in Eswatini

The impact of classroom climate on students and staff can be either beneficial or a barrier to the teaching and learning process. Unfortunately, there is no study that has been conducted on the contribution of classroom climate to the teaching of high school agriculture students in Eswatini. Therefore, the study sought to find out the contribution of classroom climate to the teaching of agriculture in senior secondary schools in Eswatini. An exploratory research design was used in the study. A total of sixteen schools were randomly selected; and one agriculture class was picked for an observation for the contribution of classroom climate to the teaching of agriculture. After the observation, an interview was conducted with the classroom teacher to complement and explain some of the observation. The observation schedule and interview protocol were peer reviewed to address issues of trustworthiness. Data were collected in February 2019. Findings revealed that the four variables: presage, context, process and products were contributing to the classroom climate; thus enhancing the teaching of agriculture. Therefore, the study recommended that agriculture must be taught by teachers who have received training on the subject matter and have the technical skill in solving problems.

Checklist to assess Trustworthiness in RAndomised Controlled Trials (TRACT checklist): concept proposal and pilot

ObjectivesTo propose a checklist that can be used to assess trustworthiness of randomized controlled trials (RCTs).DesignA screening tool was developed using the four-stage approach proposed by Moher et al. This included defining the scope, reviewing the evidence base, suggesting a list of items from piloting, and holding a consensus meeting. The initial checklist was set-up by a core group who had been involved in the assessment of problematic RCTs for several years. We piloted this in a consensus panel of several stakeholders, including health professionals, reviewers, journal editors, policymakers, researchers, and evidence-synthesis specialists. Each member was asked to score three articles with the checklist and the results were then discussed in consensus meetings.OutcomeThe Trustworthiness in RAndomised Clinical Trials (TRACT) checklist includes 19 items organised into seven domains that are applicable to every RCT: 1) Governance, 2) Author Group, 3) Plausibility of Intervention Usage, 4) Timeframe, 5) Drop-out Rates, 6) Baseline Characteristics, and 7) Outcomes.Each item can be answered as either no concerns, some concerns/no information, or major concerns. If a study is assessed and found to have a majority of items rated at a major concern level, then editors, reviewers or evidence synthesizers should consider a more thorough investigation, including assessment of original individual participant data.ConclusionsThe TRACT checklist is the first checklist developed specifically to detect trustworthiness issues in RCTs. It might help editors, publishers and researchers to screen for such issues in submitted or published RCTs in a transparent and replicable manner.

A systematic review of trustworthy and explainable artificial intelligence in healthcare: Assessment of quality, bias risk, and data fusion

In the last few years, the trend in health care of embracing artificial intelligence (AI) has dramatically changed the medical landscape. Medical centres have adopted AI applications to increase the accuracy of disease diagnosis and mitigate health risks. AI applications have changed rules and policies related to healthcare practice and work ethics. However, building trustworthy and explainable AI (XAI) in healthcare systems is still in its early stages. Specifically, the European Union has stated that AI must be human-centred and trustworthy, whereas in the healthcare sector, low methodological quality and high bias risk have become major concerns. This study endeavours to offer a systematic review of the trustworthiness and explainability of AI applications in healthcare, incorporating the assessment of quality, bias risk, and data fusion to supplement previous studies and provide more accurate and definitive findings. Likewise, 64 recent contributions on the trustworthiness of AI in healthcare from multiple databases (i.e., ScienceDirect, Scopus, Web of Science, and IEEE Xplore) were identified using a rigorous literature search method and selection criteria. The considered papers were categorised into a coherent and systematic classification including seven categories: explainable robotics, prediction, decision support, blockchain, transparency, digital health, and review. In this paper, we have presented a systematic and comprehensive analysis of earlier studies and opened the door to potential future studies by discussing in depth the challenges, motivations, and recommendations. In this study a systematic science mapping analysis in order to reorganise and summarise the results of earlier studies to address the issues of trustworthiness and objectivity was also performed. Moreover, this work has provided decisive evidence for the trustworthiness of AI in health care by presenting eight current state-of-the-art critical analyses regarding those more relevant research gaps. In addition, to the best of our knowledge, this study is the first to investigate the feasibility of utilising trustworthy and XAI applications in healthcare, by incorporating data fusion techniques and connecting various important pieces of information from available healthcare datasets and AI algorithms. The analysis of the revised contributions revealed crucial implications for academics and practitioners, and then potential methodological aspects to enhance the trustworthiness of AI applications in the medical sector were reviewed. Successively, the theoretical concept and current use of 17 XAI methods in health care were addressed. Finally, several objectives and guidelines were provided to policymakers to establish electronic health-care systems focused on achieving relevant features such as legitimacy, morality, and robustness. Several types of information fusion in healthcare were focused on in this study, including data, feature, image, decision, multimodal, hybrid, and temporal.

Trustworthy Localization With EM-Based Federated Control Scheme for IIoTs

Industrial Internet of Things (IIoTs) are significantly changing informative and manufacturing pattern in smart factories while it also brings security and trustworthiness issue. Concerning about trustworthiness issues and private preservation of tracking systems, a hierarchical framework with federated control theory is designed, which consists of a federated control center, network layer, and a federated control node. The framework combines a collaborative Cloud-Edge-End structure and machine learning-oriented localization, which further forms the EM-based federated scheme. On this basis, a trustworthy localization model is built with the untrustworthiness probability as a latent variable. By exploring expectation maximization (EM) of trustworthy localization, the local messages and aggression equations are derived in an iterative way of federated learning. The EM-based federated control scheme with machine learning-oriented localization is finally given. Experiments have been conducted to prove the localization accuracy and convergence of the proposed method with trustworthiness issue. The results show that trustworthy localization outperforms traditional methods without considering the security threats.

Trustworthiness of Laser-Induced Breakdown Spectroscopy Predictions via Simulation-based Synthetic Data Augmentation and Multitask Learning

Laser-induced breakdown spectroscopy is a versatile technique that can be used to quickly measure the concentration of elements in ambient air. We tackle the issues of performance and trustworthiness of the statistical model used for predictions. We propose a method for improving the performance and trustworthiness of statistical models for LIBS. Our method uses deep convolutional multitask learning architectures to predict the concentration of the analyte and additional information as auxiliary outputs. We also introduce a simulation-based data augmentation process to synthesize more training samples. The secondary predictions from the model are used to characterize, quantify and validate its trustworthiness, taking advantage of the mutual dependencies of the weights of the neural networks. As a consequence, these output can be used to successfully detect anomalies, such as changes in the experimental conditions, and out-of-distribution samples. Results on different types of materials show that the proposed method improves the robustness and trueness of the predictions.

Automated Economic Welfare for Everyone? Examining Barriers to Adopting Robo-Advisors from the Perspective of Explainable Artificial Intelligence

Robo-advisors (RAs) support economic decisions for customers using artificial intelligence (AI). RAs are gaining increasing significance but lack market penetration. A significant issue is the perceived low transparency of such AI systems. This study examines the public’s demands on RAs with text-mining methods from the perspective of explainable artificial intelligence (XAI). The results reveal understandability and trustworthiness issues for each of the RA use phases (configuration, matching, and maintenance). In particular, five barriers emerge in RA if information needs remain unanswered: entry barrier, assessment barrier, evaluation barrier, continuation barrier and withdrawal barrier. The barriers can be mitigated by combining explanation, design and communication measures. The results are discussed regarding theoretical implications and practical recommendations for facilitating the adoption of RAs. JEL: D8 (D81, D83, D89), G1 (G11), G2 (G20, G23), G4 (G41), I2 (I24, I25), O3 (O31, O33)

A New Dark Age? Truth, Trust, and Environmental Science

This review examines the alleged crisis of trust in environmental science and its impact on public opinion, policy decisions in the context of democratic governance, and the interaction between science and society. In an interdisciplinary manner, the review focuses on the following themes: the trustworthiness of environmental science, empirical studies on levels of trust and trust formation; social media, environmental science, and disinformation; trust in environmental governance and democracy; and co-production of knowledge and the production of trust in knowledge. The review explores both the normative issue of trustworthiness and empirical studies on how to build trust. The review does not provide any simple answers to whether trust in science is generally in decline or whether we are returning to a lessenlightened era in public life with decreased appreciation of knowledge and truth. The findings are more nuanced, showing signs of both distrust and trust in environmental science.