Secure Platform Research Articles

Enhancing a fog-oriented IoT authentication and encryption platform through deep learning-based attack detection

The term Internet of Things (IoT) refers to a network that connects smart things with sensors. Healthcare, transportation, and smart cities are some IoT applications. IoT technologies integrate objects in the cloud-based Internet. The massive scale of IoT exposes some systems to attacks. There is an urgent need for solutions that efficiently handle IoT authentication, encryption, and attack detection. This work proposes a Fog-based IoT security platform named IoTSafe. It contains mechanisms for authentication and encryption and a deep learning-based attack detection module. The IoTSafe attack detection module uses the Message Queue Telemetry Transport (MQTT). Tests were performed to evaluate the IoTSafe platform in three different environments. A case study demonstrated that the platform is efficient with all proposed mechanisms and modules. The results for the attack detection module show the proposal’s effectiveness with an accuracy of 99.57% and a precision of 99.66%. The IoTSafe time response was less than one second, guaranteeing the quality of service.

Comprehensive review of machine learning models for sql injection detection in e-commerce

With the steady expansion of online commerce, e-commerce sites have become increasingly attractive targets for hackers. These sites serve millions of customers and often hold valuable, confidential, and sometimes financial information in their databases. One particularly dangerous type of attack is SQL injection, which exploits vulnerabilities in web applications to influence backend databases, posing significant threats to such platforms. Traditional defenses like desktop firewalls, input validation, and parameterized queries provide some level of protection but are often insufficient against newer injection variations and sophisticated attackers. The utilization of machine learning to enhance cybersecurity against more advanced threats has been demonstrated as a promising approach. This systematic review examines how various machine learning algorithms are applied to detect SQL injection attacks that could potentially harm e-commerce systems. By identifying and analyzing the relevant literature, this review highlights the effectiveness of different algorithms and their practical applications in enhancing the security of online commerce platforms. More specifically, five techniques were assessed on both real and synthetic datasets: Logistic Regression, Naive Bayes, Random Forest, Artificial Neural Network, and two combined models (Logistic Regression & Naive Bayes, and Artificial Neural Network & Random Forest). The findings indicate that Random Forest performed better than other algorithms in the decision tree family, attributed to its ability to balance precision and recall effectively. However, limitations such as using a single dataset and the computational complexity of some models were noted. This review provides insights for practitioners on selecting appropriate detection models and outlines approaches to address current limitations through future work. Addressing these limitations could involve using more diverse datasets, optimizing computational efficiency, and exploring advanced ensemble methods and neural network architectures.

Secure and Privacy-Protected Bioinformation Implementation in Air Passenger Transport Based on DLT

Aviation information systems are a key component in ensuring efficient and smooth air transport operations. In this regard, the transfer of passenger information between parties is of paramount importance. With the continuous improvement of biometrics technology, this kind of individual identification that can provide accurate and unforgeable identification is widely used in various fields. This research presents the significance and effective application scenarios of facial recognition in biometrics in air transport operations. Due to the characteristics of aviation information systems, Distributed Ledger Technology (DLT) is used in this study for secure and private transmission of facial recognition information. Distributed systems can give a transparent and secure platform to multiple parties to access sensitive passenger data. This study uses the Corda framework as the DLT that supports CorDapp development. Based on the above techniques, this study proposes two feasible application scenarios. One is a baggage match detection system to prevent misplaced baggage, and the other is an iAPIS system that transmits passenger information in real-time communication between airlines and border control agencies. This article details how to apply the research in these two scenarios, as well as the benefits and implications of the applications. Finally, this article presents an outlook for future development and feasible directions for improvement.

Enhancing claustrophobia exposure therapy: A blockchain and NFT-enabled metaverse approach

Claustrophobia is a widespread mental health issue that affects a significant portion of the population. Traditional approaches, such as exposure therapies, have shown efficacy but are limited by their inability to provide immersive and controlled environments for effective therapy. These methods often struggle to replicate the intensity and authenticity of claustrophobic scenarios and face challenges in data security and patient engagement. In this paper, we propose a blockchain and Non-Fungible Token (NFT)-based metaverse to create a decentralized, transparent, immersive, and secure platform for claustrophobia exposure therapy. Our approach integrates blockchain technology for decentralized data management and NFTs for incentivizing patient participation, offering a unique enhancement to traditional therapy methods. We develop four blockchain smart contracts to manage registration, access control, NFT management, progress monitoring, and incentives distribution. Our system architecture, sequence diagrams, and algorithms are presented to demonstrate the functionality of our solution. We validate our approach through functionality testing, cost evaluation, and security analysis. Specifically, we measure the gas consumption of the smart contracts, analyze transaction costs, and assess system security against vulnerabilities. Additionally, we leverage existing literature to support the effectiveness of NFTs as incentives and the role of blockchain in enhancing data security and participation. The results show improvements in therapy engagement and data security, validating the practical implications of integrating blockchain and metaverse technologies in mental health therapy. We discuss the practical implications, challenges, and limitations of our solution, along with guidelines for generalization. The source code is publicly available on GitHub.

Bio-Inspired Hyperparameter Tuning of Federated Learning for Student Activity Recognition in Online Exam Environment

Nowadays, online examination (exam in short) platforms are becoming more popular, demanding strong security measures for digital learning environments. This includes addressing key challenges such as head pose detection and estimation, which are integral for applications like automatic face recognition, advanced surveillance systems, intuitive human–computer interfaces, and enhancing driving safety measures. The proposed work holds significant potential in enhancing the security and reliability of online exam platforms. It achieves this by accurately classifying students’ attentiveness based on distinct head poses, a novel approach that leverages advanced techniques like federated learning and deep learning models. The proposed work aims to classify students’ attentiveness with the help of different head poses. In this work, we considered five head poses: front face, down face, right face, up face, and left face. A federated learning (FL) framework with a pre-trained deep learning model (ResNet50) was used to accomplish the classification task. To classify students’ activity (behavior) in an online exam environment using the FL framework’s local client device, we considered the ResNet50 model. However, identifying the best hyperparameters in the local client ResNet50 model is challenging. Hence, in this study, we proposed two hybrid bio-inspired optimized methods, namely, Particle Swarm Optimization with Genetic Algorithm (PSOGA) and Particle Swarm Optimization with Elitist Genetic Algorithm (PSOEGA), to fine-tune the hyperparameters of the ResNet50 model. The bio-inspired optimized methods employed in the ResNet50 model will train and classify the students’ behavior in an online exam environment. The FL framework trains the client model locally and sends the updated weights to the server model. The proposed hybrid bio-inspired algorithms outperform the GA and PSO when independently used. The proposed PSOGA not only outperforms the proposed PSOEGA but also outperforms the benchmark algorithms considered for performance evaluation by giving an accuracy of 95.97%.

WITHDRAWN: A Secure and Efficient Blockchain-Powered Scheme for e-Invoicing Application

E-invoicing has become an increasingly popular form of electronic documents in B2B and B2G commerce transactions, and facilitates an effective exchange of invoice-related information containing tax invoice requirements. It enables businesses and governments to make transactions data be stored, retrieved and handled more efficiently among various aspects of business operations through an integrated systems solution such as ERP systems or e-commerce platforms. Although e-invoicing has numerous benefits, the integrated processes involve security risks associated with the ability to exchange and access critical transaction records and sensitive financial information over public networks such as the Internet. To secure e-invoicing services and maintain trust and privacy among their engaged stakeholders, many e-invoicing platforms use various authentication methods, such as factor-based mechanisms and digital signature techniques, to prevent potential threats and unauthorized access through cyberattacks. However, such authentication techniques cannot fully protect sensitive information contained in e-invoices from security breaches in the context of sharing transactional data. This proposed scheme introduces a secure authentication mechanism that integrates the self-certified methodology to enhance the trustworthiness of trading parties. In addition, the computational efforts of the proposed model show that high performance can be achieved in executing e-invoicing processes while maintaining data privacy. By leveraging the capabilities of blockchain technology and the adoption of decentralized applications and smart contracts, the proposed solution enables e-invoicing services to provide a highly secure, efficient, immutable, and transparent platform for businesses and tax authorities.

Data-Driven Web-Based Service Travel Resume Information System

The Energy and Mineral Resources Department (ESDM) of South Sumatra Province has significantly enhanced its operational efficiency by implementinga Web-Based Official Travel Resume Information System. This system was developed to address the inefficiencies and risks associated with the previous manual method of managing official travel data. The new system leverages PHP, MySQL, Javascript, JQuery, and Bootstrap to provide a robust, secure, and user-friendly platform for handling travel reports, assignment orders, and official travel documentation. The digital transformation has streamlined data entry, improved data integrity, and facilitated real-time information access,enabling better decision-making and strategic planning. A System Usability Scale (SUS) test was conducted with 120 participants to evaluate the system's usability. The results showed a mean SUS score of 78.5, indicating good usability, with a standard deviation of 8.3. High scores were recorded for the system's ease of use, integration of functions, and user confidence, while lower scores highlighted initial complexity and a need for additional user training. The successful implementation and positive usability results underscore the system's effectiveness in enhancing the ESDM Department's operational capabilities. Recommendations for further improvements include enhancing mobile accessibility, providing ongoing user training, integrating advanced reporting features, and refining the user interface to ensure continued user satisfaction and system efficiency.

Communication security of intelligent information service platform combining AES and ECC algorithms

ABSTRACT The rapid progress of information technology has made the status of information service platforms increasingly prominent in social operations. However, their communication security is also facing unprecedented challenges. In response to the issues of insufficient security and efficiency in traditional encryption algorithms, this study combines advanced encryption standard algorithms with elliptic curve cryptography algorithms to generate encryption key pairs through elliptic curve cryptography algorithms. It utilizes advanced encryption standard algorithms to generate symmetric keys, and then encrypts the keys of the advanced encryption standard using the public key of the elliptic curve cryptography algorithm. Based on this, it implements advanced encryption standard encryption on data and designs a hybrid encryption algorithm. The results showed that the throughput of this method was 4000bit/s, with a delay of 550 m, which is higher than other algorithms, proving its high efficiency in processing data. The average bit change of the ciphertext in the hybrid encryption algorithm was 55.8 bits, with a range of bit changes of [69.3, 82.7], which is significantly higher than other algorithms, proving its high security. The designed algorithm has significant benefits in communication security of information service platforms and plays an important role in promoting the development of future network security defense technology.

Cloud-Based Transaction Fraud Detection: An In-depth Analysis of ML Algorithms

Context: Cloud-based services are increasingly central in financial technology, enabling scalable and efficient transactions. However, they also heighten vulnerability to fraud, challenging the security of online financial activities. Traditional fraud detection struggles against sophisticated tactics, highlighting the need for advanced, cloud-compatible solutions. Objectives: This study assesses machine learning (ML) algorithms' ability to detect fraud in cloud environments, focusing on Logistic Regression (LR), Decision Trees (DT), Random Forest (RF), Support Vector Machines (SVM), and XGBoost (XGB). It uses a comprehensive dataset to determine which ML model best identifies fraudulent transactions, aiming to optimize these models for accuracy, precision, and efficiency in real-time detection. Results: The XGBoost model outperformed others in fraud detection accuracy, with Random Forest and Decision Trees also showing high effectiveness. These models were particularly good at balancing precision and recall, minimizing false positives, and accurately identifying fraud in complex transaction patterns. Conclusion: ML, especially ensemble and boosting models like XGBoost and Random Forest, offers a strong approach to detecting transaction fraud in cloud-based financial systems. Their capacity to handle vast data volumes and adapt to new fraud patterns enhances cloud transaction security. Implication: The study provides a guide for financial and cloud services to implement sophisticated fraud detection systems. It emphasizes the importance of continual ML innovation to tackle digital finance fraud, suggesting that adopting these advanced ML models can significantly reduce transaction fraud risks, ensuring a secure, efficient, and trustworthy platform for users.

Morpheus: An efficient timing-based attestation framework for safeguarding hypervisor integrity with dynamic trust

Hypervisor, the core software in cloud computing, is susceptible to malicious intrusions, potentially jeopardizing overall platform security. Physical hosts within the cloud computing environment constantly face persistent attacks, and not all hosts are equipped with essential security hardware. The prolonged latency of previous software-based detection methods could not comprehensively address these cloud threats. Furthermore, the widespread deployment of security hardware incurs substantial costs for hardware-based detection. To address these challenges, we introduce Morpheus, an efficient framework ensuring hypervisor integrity in the current cloud. This framework rapidly employs software-based methods to detect malicious hosts, utilizing a subset of hosts equipped with security hardware as the Root of Trust. Efficiency is augmented through a Neural Network scheduling module, and an embedded exponential aging mechanism fortifies time-aging trust against consistent cloud threats. Evaluation results demonstrate that it can promptly identify threatened hosts with acceptable system overhead loss, solidifying its position as a robust cloud security solution.

Authenticated Range Querying of Historical Blockchain Healthcare Data Using Authenticated Multi-Version Index

With growing adoption of blockchain in established and emerging applications, there is an increasing need to support efficient ad hoc querying of authenticated historical data. This is especially true in fields such as healthcare to meet the rigorous security and regulatory requirements of ever-expanding digital health platforms. Existing blockchain systems, however, offer little or no support for querying capabilities over historical data. Although a full blockchain archive node can be used to maintain historical records of all executed transactions on the chain, it is not scalable when dealing with large volumes of data. Moreover, such ‘offline’ historical data lack tamper evidence support. To address these issues, we introduce an authenticated index structure called Authenticated Multi-Version Skip List (AMVSL), designed to support a rich set of query features over historical blockchain data. We further present three range queries: SVRK, MVRK, and MVAK, which offer querying over a range of keys and a range of versions. Our experimental evaluation of two healthcare-inspired examples demonstrates that AMVSL efficiently supports these queries and can achieve performance that is several orders of magnitude faster than existing authenticated data structures.

Development of Cyber Security Platform for Experiential Learning

Development of Cyber Security Platform for Experiential Learning

Leveraging artificial intelligence for enhanced cybersecurity: insights and innovations

In the digital age, cyber threats have become increasingly sophisticated, necessitating innovative approaches to bolster security measures. Artificial Intelligence (AI) has emerged as a formidable tool in the realm of cyber security, offering advanced capabilities in threat detection, anomaly detection, and response automation. This article provides an overview of AI applications in cyber security, highlighting its role in mitigating risks and fortifying defense mechanisms. AI techniques such as machine learning, deep learning, and natural language processing empower security systems to analyze vast amounts of data in real-time, identifying patterns indicative of malicious activities. Through the utilization of AI-driven algorithms, cyber security platforms can proactively detect and neutralize cyber threats before they inflict substantial damage. Moreover, AI enables the automation of incident response processes, reducing response times and minimizing the impact of security breaches. Case studies from leading cyber security firms from the integral part of the studies and demonstrate the practical implementation of AI-driven solutions in safeguarding critical infrastructures against cyber threats. Resilience towards cyber-attacks and safeguarding sensitive data assets through leveraging AI technologies has been focused in the study.

Сучасні методи автентифікації та авторизації користувачів розподілених веборієнтованих систем

The analysis was carried out and the results of a comparative review of the most common methods of authentication and authorization of users of web-oriented systems with a distributed architecture were presented. Considering the relevance of cybersecurity issues in the digital age, the research focuses on identifying effective strategies for protecting user data in the development of distributed web-oriented systems within the trade sector. The most likely threats to data access, characteristic of distributed web-based systems, have been studied, and the potential causes of these vulnerabilities have been determined. Particular attention in the publication is devoted to assessing the risks and benefits of various approaches, including basic authentication, session-based authentication, JWT tokens, and access and refresh tokens (OAuth 2.0 standard). Various aspects of each method have been analyzed, particularly their reliability and vulnerability to attacks. The work discusses real cases of vulnerabilities in distributed web-oriented systems and offer recommendations for their elimination to enhance the security of online trading platforms.

Ethical IT Decision Making and Data Governance

This paper explores the importance of data governance in influencing ethical decision-making in the context of information technology (IT). Data governance is used as a management technique to ensure the quality of organisational data by establishing and implementing policies, procedures and standards. Three main types of data governance are discussed: silo-oriented data governance, functional data governance, and platform data governance. The impact of each type on ethical IT decision-making is analysed, highlighting the challenges and benefits associated with each approach. Ethical IT decision-making hinges on robust data governance. Silo-oriented governance fosters fragmentation, conflicting standards, and security vulnerabilities. Functional Data Governance tailors policies to departments, enhancing accountability and risk mitigation but may lack unified ethics. Platform Data Governance builds on this, ensuring clearer accountability, data privacy, and unified ethical standards. To foster ethical IT decision-making, prioritize Platform Data Governance, establish a data platform for security, cultivate a unified ethical culture, and enforce organizational standards. These steps promote accountability, mitigate risks, and safeguard data integrity, crucial for ethically sound IT decisions in today's complex landscape.

Ethical IT Decision Making and Data Governance

This paper explores the importance of data governance in influencing ethical decision-making in the context of information technology (IT). Data governance is used as a management technique to ensure the quality of organisational data by establishing and implementing policies, procedures and standards. Three main types of data governance are discussed: silo-oriented data governance, functional data governance, and platform data governance. The impact of each type on ethical IT decision-making is analysed, highlighting the challenges and benefits associated with each approach. Ethical IT decision-making hinges on robust data governance. Silo-oriented governance fosters fragmentation, conflicting standards, and security vulnerabilities. Functional Data Governance tailors policies to departments, enhancing accountability and risk mitigation but may lack unified ethics. Platform Data Governance builds on this, ensuring clearer accountability, data privacy, and unified ethical standards. To foster ethical IT decision-making, prioritize Platform Data Governance, establish a data platform for security, cultivate a unified ethical culture, and enforce organizational standards. These steps promote accountability, mitigate risks, and safeguard data integrity, crucial for ethically sound IT decisions in today's complex landscape.

User-informed tailoring of electronic and in-person peer navigation strategies to improve retention in HIV care for adolescents and young adults: a qualitative inquiry

ABSTRACT Poor adherence and retention in HIV care remain a major challenge among adolescents and young adults (AYA) living with HIV in sub-Saharan Africa (SSA). Strategies are urgently required to support AYA to remain in care for better health outcomes. We explored AYA preferences regarding the format and delivery of electronic and in-person peer navigation to improve HIV care outcomes. This formative qualitative study was conducted among AYA enrolled in HIV care at three clinics in western Kenya. We conducted two focus group discussions (FGDs) each with 8–9 participants (n = 17) purposively selected based on age, gender and clinic where they received care. The characteristics desired of a navigator are a person of the same age group and HIV status who has a good memory and is friendly and able to maintain confidentiality. AYA want the content of their interaction with the navigator to center on sharing motivational messages and also educating them on matters of HIV care, sexual and reproductive health and mental health. The preferred navigation formats for electronic communication are platforms considered confidential. AYA preferred interventions delivered through secure communication platforms by navigators with whom they have commonalities. The navigation interventions that prioritize confidentiality and holistic content will likely be most highly valued by AYA. Furthermore, electronic mechanisms can help support the relationship building that is at the core of our navigation approach and a fundamental aspect of social work in general.

The Telehealth chain: a framework for secure and transparent telemedicine transactions on the blockchain.

Blockchain technology provides a secure and decentralized platform for storing and transferring sensitive medical data, which can be utilized to enable remote medical consultations. A theoretical framework for creating a blockchain-based digital system created to facilitate telemedicine system. This paper proposes a theoretical framework based on Hyperledger fabric for creating a blockchain-based digital entity to facilitate telemedicine services. The proposed framework utilizes blockchain technology to provide a secure and reliable platform for medical practitioners to interact remotely with patient transactions. The blockchain will serve as a one-stop digital service to secure patient data, ensure privacy, and facilitate payments. The proposed framework leverages the existing Hyperledger fabric platform to build a secure blockchain-assisted telemedicine platform.

Rapid survey assessment: Barriers and facilitators to mammogram completion in English and Spanish-speaking patients in a federally qualified health center (FQHC) setting.

1587 Background: Mammography has decreased breast cancer mortality in the US by 40% since the 1980s; yet disparities in screening and outcomes persist in underserved populations. Many studies note language as a barrier to screening with lower mammogram completion amongst Spanish-speaking patients versus those with English language proficiency. At our health center, Spanish-speaking patients were more likely to complete screening mammography compared to patients with English proficiency (53% vs 31% in 2022). To understand this pattern and barriers and facilitators to breast cancer screening, we conducted a mixed methods rapid assessment survey as part of a quality improvement initiative. Methods: We developed a patient bilingual survey measuring knowledge, attitudes, and behaviors about breast cancer prevention and screening. We identified consecutive female patients (age 40-75 years) and extracted baseline demographic data from the electronic medical record. Surveys were distributed via our health center’s patient messaging service and responses stored on a secure platform. Data analysis included descriptive statistics, chi-squared testing between groups and qualitative analysis to identify common themes from free-text answers. Results: The overall response rate was 14% (N=732/5,332) in 30 days. Of the surveys completed, 37% were in Spanish. Spanish-speaking patients more strongly regarded mammography as safe and effective compared to English-speaking patients (p<0.001). Spanish-speaking patients were more likely to view getting a mammogram as a priority (79% vs 65% p<0.001) and noted speaking about breast cancer prevention in their communities more often (p<0.006) as an important factor. Spanish-speaking patients primarily cited cost, scheduling, and language (p<0.05) as barriers to mammogram completion, while English-speaking patients cited time constraints, transportation, and discomfort (p<0.05) more frequently. Spanish-speaking patients were more likely to prefer speaking with a trusted community member or attending health fairs than English-speaking patients (P<0.05). Conclusions: Our study demonstrates differences in attitudes, behaviors, and perceived barriers to mammography in English vs Spanish-speaking patients. The increased mammogram completion rate in Spanish-speaking patients may reflect more positive attitudes about screening and more discussion in the community amongst Spanish-speaking patients queried. Reported barriers to screening differed and will guide targeted QI interventions. We demonstrated feasibility and patient engagement in a diverse population using a message system built into our EHR in an FQHC setting. This survey study allows us to elevate patient voices, better understand gaps and preferences, and better tailor future interventions for cancer screening.

Human-AI teams to improve accuracy and timeliness of oncology trial prescreening: Preplanned interim analysis of a randomized trial.

1524 Background: Identifying eligible patients for oncology clinical trials (“prescreening”) relies on manual chart review by clinical research coordinators (CRCs), which is time-consuming and often inaccurate. Consequently, 70% of patients with cancer who meet trial eligibility criteria are not offered participation. Natural language processing (NLP) may improve the accuracy and timeliness of prescreening. Methods: This was a preplanned interim analysis of a paired-design noninferiority trial, powered to assess timeliness. We adapted NLP algorithms to identify 13 common eligibility criteria related to cancer type/stage, prior systemic therapy, actionable biomarkers, and response criteria. NLP systems performed optical character recognition, entity and relationship extraction, and natural language inference through deep learning and symbolic AI techniques. Deidentified unstructured electronic health records (EHRs) from real-world patients with non-small cell lung (NSCLC) or colorectal (CrCa) cancer were presented to CRCs via a secure platform. Two CRCs were randomized 1:1 to view blocks of 20 charts each with (Human+AI) or without (Human-alone) NLP annotations. In pre-trial assessment, the CRCs had 86% inter-rater agreement. The primary outcome was overall chart-level accuracy, defined as percent of the 13 CRC-coded eligibility items matching a gold standard set, determined by 2-3 clinicians blinded to experimental arms. Paired t-tests compared overall accuracy (noninferiority margin 5%) and criteria-specific accuracy. Significant differences were assessed for superiority using two-sided paired t-tests. Mann-Whitney tests compared the secondary outcome of timeliness (time per chart review). Results: Among 74 (40 NSCLC; 34 CrCa) of a planned 400 patients, overall accuracy for Human+AI was noninferior to Human-alone (78.7% vs. 76.7%, mean difference 2.0%, p<0.001 rejecting inferiority); both were greater than AI-alone (63.5%). Superiority for Human+AI was demonstrated for RECIST response, but not for overall nor other criteria-specific accuracy (Table). Median time per review was lower for Human+AI than Human-alone (34.1 vs 43.9 min, adjusted p=0.05). Conclusions: Human+AI teams can improve timeliness of trial prescreening with noninferior accuracy. This platform is being used for eligibility assessment in an ongoing clinical trial. [Table: see text]