IEEE Open Journal of the Computer Society最新文献

{"title":"Integrating AI and Large Language Models for Automated Data Quality Enhancement in Data Integration Systems","authors":"Nidhin Karunakaran Ponon;Maria Anurag Reddy Basani","doi":"10.1109/OJCS.2026.3666345","DOIUrl":"https://doi.org/10.1109/OJCS.2026.3666345","url":null,"abstract":"This article introduces an AI and LLM-based framework to automate data quality improvement in complex data systems. Traditional methods struggle with semantic inconsistencies and evolving schemas, degrading quality as data scales. The framework incorporates Real-Time Semantic Annotation (RTSA), adaptive ontology reinforcement, contextual similarity for duplicate detection, and continuous auto-healing. Explainability is ensured via SHAP-based alignment for transparency. Evaluated on the GOBY Benchmark dataset, it achieved 89.4% semantic annotation accuracy, outperforming the strongest baseline by 3%. The duplicate reduction rate was 64.5%, and the quality score averaged 83.2%, validating the auto-healing loop's effectiveness. It adapts to evolving data without retraining, confirmed by robust performance under semantic drift. The explainability analysis showed a low SHAP divergence of 0.11, aligning model predictions with feature importance. An ablation study confirmed that the semantic feedback loop and ontology reinforcement significantly contribute to stability. This research offers a scalable solution for enhancing data quality by integrating AI-driven adaptability and LLM-based semantic understanding with built-in explainability for enterprise-grade data pipelines.","PeriodicalId":13205,"journal":{"name":"IEEE Open Journal of the Computer Society","volume":"7 ","pages":"528-540"},"PeriodicalIF":0.0,"publicationDate":"2026-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11399888","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147362456","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Advancing Autism Spectrum Disorder Diagnosis With Multimodal Data: A Survey","authors":"K. Namitha;S. Girish;M. P. Anuvind;R. S. Harish Kumar;Harishankar Binu Nair;Dhanya Chandran","doi":"10.1109/OJCS.2026.3666102","DOIUrl":"https://doi.org/10.1109/OJCS.2026.3666102","url":null,"abstract":"Autism Spectrum Disorder (ASD) is a complicated neurodevelopmental condition characterised by a variety of behavioural and communication challenges, marking timely and accurate diagnosis and intervention essential. Despite their widespread occurrence, conventional diagnostic methods like clinical observations and standardised questionnaires are hindered by issues like subjectivity, practitioner variability, and long evaluation durations and existing works using AI for diagnosis falters as they fail to capture range of symptoms, generalizability across cohorts and a lot more challenges. This survey compiles a comprehensive set of ASD datasets, including the limited multimodal collections, and outlines clear modality specific preprocessing steps that ensures data quality. By reviewing prior multimodal fusion surveys and highlighting breakthroughs across individual modalities with the main fusion studies, it establishes a unified view of the diagnostic approach and the fusion techniques used in ASD research. Furthermore, the survey explains the advantages of multimodal fusion, summarises the major limitations reported across existing work, and introduces a simple taxonomy with practical solutions that address these challenges and guide future development.","PeriodicalId":13205,"journal":{"name":"IEEE Open Journal of the Computer Society","volume":"7 ","pages":"541-559"},"PeriodicalIF":0.0,"publicationDate":"2026-02-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11398113","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147362454","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Reinforcing CBDC Integrity: A Novel Anti Money Laundering Solution by Integrating Blockchain, Machine Learning, and Taint Analysis","authors":"Shamama Tul Amber;Huma Ghafoor;Pham Duy Thanh;Chih-Hsien Hsia","doi":"10.1109/OJCS.2026.3666199","DOIUrl":"https://doi.org/10.1109/OJCS.2026.3666199","url":null,"abstract":"As the central bank digital currency (CBDC) is rapidly becoming operational, there are both opportunities for financial inclusion and efficiency, as well as problems of security threats and anti-money laundering (AML), particularly due to concerns about user anonymity. In order to address the risks associated with anonymous transactions, this research proposes a multi-layered security framework for CBDC that makes use of smart contracts built on the Ethereum platform, taint analysis, and machine learning (ML) models. The solution makes use of programmable smart contracts to automate policy enforcement and transaction validation within the CBDC ecosystem. Taint analysis techniques are incorporated to track the movement of illicit funds and identify questionable transaction patterns across the blockchain network. This is further enhanced by ML models that are optimized to learn from transaction data in order to reliably identify anomalous or illicit actions before they occur. We have generated two synthetic datasets that include two case scenarios and trained six ML models to evaluate them comparatively. Of them, random forest had the highest level of accuracy, 91.11%, in the cross-border case, whereas the support vector machine had a accuracy of 93.12% in the case of real estate transactions. In addition, we conducted a performance comparison of five environments; traditional banking, CBDC with baseline blockchain, CBDC with blockchain, CBDC with ML, and CBDC with blockchain, taint analysis and ML. We used different metrics to test the performance of our proposed scheme and found that our AML tracking algorithm took an average of 0.55 s inference time, which is faster than the underlying reference method. According to our results, the proposed combined framework ensures high-level protection that improved risk detection in digital currencies, with a significantly reduced risk of money laundering and related hazards when using CBDC systems.","PeriodicalId":13205,"journal":{"name":"IEEE Open Journal of the Computer Society","volume":"7 ","pages":"514-527"},"PeriodicalIF":0.0,"publicationDate":"2026-02-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11399631","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147299667","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"MCP-Secure: A Runtime Access Control Layer for Privilege-Aware LLM Agent Tooling","authors":"Gamini Singh;Vijay K. Madisetti","doi":"10.1109/OJCS.2026.3664314","DOIUrl":"https://doi.org/10.1109/OJCS.2026.3664314","url":null,"abstract":"Tool-enabled language-model agents introduce new security risks because their behavior evolves over multi-step workflows, yet existing defenses primarily rely on static allowlists or infrastructure isolation. This article presents MCP-Secure, a lightweight, host-side enforcement layer for the Model Context Protocol (MCP) that applies scoped access, read-only defaults, and approval-gated privilege elevation at runtime. MCP-Secure tracks permissions through a session-level state machine and mediates every tool invocation without modifying agents or MCP servers. We evaluate the framework across 1,080 executions spanning multiple models, tasks, and adversarial simulations. Results show that scoped access alone blocks most unsafe behaviors, read-only enforcement reliably neutralizes all mutating attack vectors, and approval-gated elevation maintains strong safety while enabling controlled write operations. Across configurations, the wrapper also shapes agent planning, reducing unsafe attempts as policies tighten. These findings demonstrate that MCP-Secure provides a practical, reproducible mechanism for enforcing least-privilege constraints in tool-enabled LLM systems, offering strong adversarial resistance with interpretable safety– utility trade-offs.","PeriodicalId":13205,"journal":{"name":"IEEE Open Journal of the Computer Society","volume":"7 ","pages":"504-513"},"PeriodicalIF":0.0,"publicationDate":"2026-02-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11394790","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147299668","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Risk Model for Choosing Signature Thresholds for Software Updates","authors":"Brian Romansky;Thomas Mazzuchi;Shahram Sarkani","doi":"10.1109/OJCS.2026.3663493","DOIUrl":"https://doi.org/10.1109/OJCS.2026.3663493","url":null,"abstract":"The timely installation of software updates has become an accepted best practice for protecting connected systems from persistent threats. Modern software update frameworks and trust anchor management protocols allow for the use of multiple digital signatures on update messages to protect against the potential compromise or loss of a digital signature private key or compromise of the underlying digital signature platform. The choice of whether to use multiple signatures and the specific number of signatures to use should be based on a risk assessment. We have developed a probability model that provides relative risk metrics for several different policy and system design options that include multiple signatures. The model enables system architects and operators to make a risk-based decision when considering multiple signatures to protect critical software updates.","PeriodicalId":13205,"journal":{"name":"IEEE Open Journal of the Computer Society","volume":"7 ","pages":"480-491"},"PeriodicalIF":0.0,"publicationDate":"2026-02-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11389194","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147299670","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Vision-Based Framework for Safe Landing Zone Mapping of UAVs in Dynamic Environments","authors":"Sakineh Abdollahzadeh;Mohand Saïd Allili;Aissa Boulmerka;Jean-François Lapointe","doi":"10.1109/OJCS.2026.3663268","DOIUrl":"https://doi.org/10.1109/OJCS.2026.3663268","url":null,"abstract":"Identification safe landing zones (SLZ) for Uncrewed Aerial Vehicles (UAVs) is important to ensure reliable and safe navigation, especially when they are operated in complex and safety-critical environments. However, this is a challenging task due to obstacles and UAV motion. This paper proposes a vision-based framework that maps SLZs in dynamic scenes by integrating several functionalities for analyzing visually static and dynamic aspects of a scene. Static analysis is achieved through context-aware segmentation which divides the image into thematic classes enabling to identify suitable landing surfaces (e.g., roads, grass). For dynamic content analysis, we combine object detection, tracking, and trajectory prediction to determine object occupancy and identify regions free of obstacles. Trajectory prediction is performed through a novel encoder–decoder architecture taking past object positions to predict the most likely future locations. To ensure stable and robust trajectory prediction, we introduce an optimized homography computation using multi-scale image analysis and cumulative updates to compensate UAV motion. We tested our framework on different operational scenarios, including urban and natural scenes with moving objects like vehicles and pedestrians. Obtained results demonstrate its strong performance, and its significant potential for enabling autonomous and safe UAV navigation.","PeriodicalId":13205,"journal":{"name":"IEEE Open Journal of the Computer Society","volume":"7 ","pages":"492-503"},"PeriodicalIF":0.0,"publicationDate":"2026-02-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11387746","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147299675","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"DHT-Backed Ancestor-Assisted Merkle Verification for Scalable and Real-Time Log Integrity in Cloud Data Lakes","authors":"Vayuphak Saengthong;Theme Mantharngkul;Chatdanai Wongsuwan;Korarit Ritthaisong;Somchart Fugkaew","doi":"10.1109/OJCS.2026.3663463","DOIUrl":"https://doi.org/10.1109/OJCS.2026.3663463","url":null,"abstract":"Cloud data lakes accumulate massive volumes of append-only, high-frequency logs from web servers, gateways, security appliances, and databases. Ensuring their integrity is challenging because existing blockchain-assisted auditing schemes are primarily optimized for static or periodically updated data and provide limited support for continuous ingestion, real-time verification, and auditor access control. These limitations lead to high recomputation overhead, delayed tamper detection, and the lack of reusable proofs required for scalable log publishing. This article introduces a verifiable log publishing framework based on a newly proposed ancestor-assisted Merkle Tree with stride-based checkpoints for efficient verification of encrypted logs. Each log record is canonically formatted, AEAD-encrypted with metadata binding, and producer-attested prior to storage. Parent-node records and ancestor certificates are distributed via a private Distributed Hash Table (DHT) to enable proof reuse, while only an epoch-level root-of-roots is anchored on a public blockchain to provide tamper-evident integrity guarantees. The framework further supports auditor authorization using certification-authority-issued tokens and incorporates micro-root commitments for near real-time verification. Experimental results demonstrate substantial performance improvements: single-record verification is significantly faster than the strongest baseline, verification throughput is markedly higher, commit latency is greatly reduced, and on-chain storage remains constant across all evaluated epoch sizes. These results confirm that the proposed architecture enables scalable, low-latency, and compliance-aware log integrity verification suitable for modern cloud data lake environments.","PeriodicalId":13205,"journal":{"name":"IEEE Open Journal of the Computer Society","volume":"7 ","pages":"468-479"},"PeriodicalIF":0.0,"publicationDate":"2026-02-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11389188","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147299671","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Scalability and Sustainability of Linear WSNs Using Erasure Coding-Based TDMA Scheduling","authors":"Nguyen Viet Ha;Tran Thi Thao Nguyen;Masato Tsuru","doi":"10.1109/OJCS.2026.3661540","DOIUrl":"https://doi.org/10.1109/OJCS.2026.3661540","url":null,"abstract":"Linear multi-hop wireless sensor networks (LWSNs) are commonly used for infrastructure monitoring and event detection where single-hop communication and wired links are impractical. Despite their cost-effectiveness and ease of deployment, these networks often suffer packet loss due to link instability, interference, and varying traffic loads. This article presents a proactive packet loss recovery method combining hop-by-hop Erasure Coding (EC) with Time Division Multiple Access (TDMA) scheduling and investigates its scalability and sustainability. EC encodes multiple packets into linear combinations, enabling decoding as long as the number of received coded packets meets or exceeds the number of original packets. This approach enhances robustness against unpredictable losses and fluctuating data rates. We present a general EC-based TDMA timeslot scheduling model for LWSNs of any size, supporting concurrent data uplink and control downlink delivery within one cycle. A heuristic search algorithm efficiently finds near-optimal timeslot allocations, meeting energy consumption constraints while maximizing the successful delivery probability. Simulations on a 15–15 topology with a high link loss rate of 0.5, where each node generates 5 data packets per minute, achieve a successful delivery probability exceeding 0.99, while power analysis shows that nodes can sustain 10 days of operation with a 1000 <italic>mAh</i>, 3.2 <italic>V</i> LiFePO<inline-formula><tex-math>$_{4}$</tex-math></inline-formula> battery. These results demonstrate that the proposed EC-based TDMA scheduling is a reliable, energy-efficient, and practical solution for real-world LWSN deployments.","PeriodicalId":13205,"journal":{"name":"IEEE Open Journal of the Computer Society","volume":"7 ","pages":"430-442"},"PeriodicalIF":0.0,"publicationDate":"2026-02-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11372907","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146223806","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"zk-REAL: A Zero-Knowledge-Based Protocol for Repeated Image Edit Authenticity Proof With Lattice Hashing","authors":"Atsuki Koyama;Kentaroh Toyoda;Manato Fujimoto;Thi Hong Tran","doi":"10.1109/OJCS.2026.3661052","DOIUrl":"https://doi.org/10.1109/OJCS.2026.3661052","url":null,"abstract":"The rapid advancement of deepfake technology poses risks, including financial fraud and political misinformation, demanding robust methods for verifying image content authenticity. While C2PA and ZK-based methods provide mechanisms for authenticity, existing solutions struggle to efficiently support privacy-preserving edits and iterative modifications. We present zk-REAL, a system that combines lattice-based hashing (LHF) with a Halo2 circuit design to reduce constraint cost and to exploit LHF updatability for partial recomputation after local edits. To remain interoperable with standard toolchains, a SHA-256 hash is computed outside the circuit as a compatibility layer with C2PA. On 720p images, our implementation demonstrates reduced key-generation cost (2× faster) and proving time for complex transformations. For repeated edits we achieve up to 29% proving-time reduction relative to recomputing from scratch, and 48% for double-edit contrast. Trade-offs include larger proof sizes for simple operations (crop, resize) versus smaller proofs for complex operations (contrast), indicating suitability for computationally intensive multi-edit workflows.","PeriodicalId":13205,"journal":{"name":"IEEE Open Journal of the Computer Society","volume":"7 ","pages":"443-454"},"PeriodicalIF":0.0,"publicationDate":"2026-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11372060","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147299666","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"ECommVis: Supporting E-Commerce Marketplace Advertising Outcomes Through a Visual Analytics System","authors":"Henrique L. S. Gino;Jean R. Ponciano;Claudio D. G. Linhares;Agma J. M. Traina","doi":"10.1109/OJCS.2026.3660917","DOIUrl":"https://doi.org/10.1109/OJCS.2026.3660917","url":null,"abstract":"E-commerce marketplace platforms rely increasingly on digital advertising to drive merchant sales, yet existing analytics tools often fail to provide contextual insights that connect advertising performance with actionable business strategies. This paper introduces ECommVis, a visual analytics platform designed to address the gap between complex advertising metrics and merchant decision-making in e-commerce marketplaces. ECommVis integrates five complementary visualization components that provide temporal, spatial, and comparative perspectives on advertising performance: Orders Over Time (OS) tracks historical organic versus ad-driven sales through adjustable time frames; Geospatial Interactions (GS) visualizes spatial performance patterns using hexagonal grid mapping; Merchant Metrics (MC) displays advertising KPIs via interactive stacked charts correlating budget with engagement; Interactions by Weekday and Shift (HM) identifies optimal advertising windows across daily shifts and weekday patterns; and Time Period Comparison (TP) benchmarks campaigns across time frames using correlation, trend, and search term analyses. We demonstrate the usefulness and effectiveness of ECommVis through two usage scenarios and a user study with 11 participants.","PeriodicalId":13205,"journal":{"name":"IEEE Open Journal of the Computer Society","volume":"7 ","pages":"455-467"},"PeriodicalIF":0.0,"publicationDate":"2026-02-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11370699","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147299669","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}