Expert Systems with Applications最新文献

{"title":"An integrated space polyhedral grid grey relational analysis model based on panel interval grey number for seawater quality assessment","authors":"Xuemei Li ,&nbsp;Zhichao Chen ,&nbsp;Yufeng Zhao ,&nbsp;Shiwei Zhou","doi":"10.1016/j.eswa.2025.127363","DOIUrl":"10.1016/j.eswa.2025.127363","url":null,"abstract":"<div><div>Deterministic models struggle to fully capture the fluctuations, trends, and interactions among various factors in complex systems. Moreover, existing panel data grey relational models are inadequate for simultaneously addressing the relationships between interval grey numbers and real numbers. To address this issue and depict uncertain information, this study proposes an interval grey number space polyhedral grid grey relational analysis model. Specifically, this model employs panel interval grey numbers as descriptors and fully exploits the grey information they contain by constructing a spatial double tetrahedral grid structure, thereby enhancing accuracy. Additionally, a binary index behavior sub-matrix is introduced to mitigate the impact of sample sequence changes on the fluctuations of grey relational degrees. The proposed model was comprehensively applied to analyze the generating and driving factors of China’s seawater quality and compared with other grey correlation models. Results indicate that the volume of sewage discharged from direct marine pollution sources is the most significant driving factor affecting China’s seawater quality, while eutrophication in coastal waters is the primary generating factor for changes in seawater quality. This study not only establishes a connection between IGN and real numbers but also enhances grey relational analysis from a space perspective. The findings offer robust theoretical support and practical guidance for applying the grey relational model in marine environmental protection and governance.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"277 ","pages":"Article 127363"},"PeriodicalIF":7.5,"publicationDate":"2025-03-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143697239","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Autonomous in-situ modeling for virtual building models in digital twins","authors":"Jeyoon Lee ,&nbsp;Jiteng Li ,&nbsp;Sungmin Yoon","doi":"10.1016/j.eswa.2025.127289","DOIUrl":"10.1016/j.eswa.2025.127289","url":null,"abstract":"<div><div>A virtual building model (VBM) is a mathematical representation that serves as a virtual replica, depicting the physical behavior of real building systems in a building digital twin (DT). DT-enabled building operations leveraging VBMs have contributed to reducing energy consumption during building operations. In-situ modeling approaches strive to develop accurate VBMs; however, the autonomous development of VBMs using these approaches remains unexplored. To address this gap, this study introduces an autonomous in-situ modeling (AIM) method, representing the first attempt to autonomously develop an unobserved virtual model. AIM aims to calibrate pre-built models using autonomously developed virtual models. This method was applied to predict evaporator inlet temperatures in a real operating heating, ventilating, and air conditioning (HVAC) system, as well as the secondary-side supply water pressure in a district heating substation (DHS) system. The effectiveness of the AIM-developed virtual evaporator inlet temperature model was evaluated by comparing the results with expert-developed virtual models. The AIM-driven target virtual model achieved a root mean squared error (RMSE) accuracy of 0.40 °C, compared to 0.73 °C when expert intervention was involved. The AIM-driven virtual secondary-side supply water pressure model achieved an RMSE of 0.16 Pa, indicating that AIM can serve as a generalizable algorithm applicable to various building energy systems. These results validate AIM’s feasibility and highlight its potential to extend the virtual model infrastructure within a DT environment. This advancement paves the way for DT-enabled autonomous building operation.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"278 ","pages":"Article 127289"},"PeriodicalIF":7.5,"publicationDate":"2025-03-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143748507","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Improved fuzzy control charts for monitoring defined health ranges using trapezoidal fuzzy numbers","authors":"Muhammad Usman Aslam ,&nbsp;SongHua Xu ,&nbsp;Zahid Rasheed ,&nbsp;Muhammad Noor-ul-Amin ,&nbsp;Sajid Hussain ,&nbsp;Muhammad Waqas","doi":"10.1016/j.eswa.2025.127310","DOIUrl":"10.1016/j.eswa.2025.127310","url":null,"abstract":"<div><div>Healthcare monitoring requires precise and efficient methods to monitor individual health measurements, particularly for diseases with well-defined clinical ranges. Traditional control charts struggle to handle uncertainty in medical data, necessitating more flexible approaches. This study introduces two novel fuzzy control charts: the fuzzy moving average control chart (FMACC) and the fuzzy weighted moving average control chart (FWMACC), which utilize trapezoidal fuzzy numbers (TrFNs) to enhance monitoring capabilities. An α-cut midrange approach is applied to better capture variability, and fuzzy process capability indices (FPCIs) are incorporated to assess process performance under uncertain conditions. The proposed method is applied to creatinine and PCR data, demonstrating its versatility in health monitoring. Monte Carlo simulations validate the effectiveness of FMACC and FWMACC, confirming their superior performance in detecting small process shifts. The findings highlight the effectiveness of proposed control charts for healthcare applications, offering a significant advancement in statistical process monitoring by integrating fuzzy logic. This approach provides a robust tool for healthcare professionals to monitor patient data more reliably and efficiently.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"278 ","pages":"Article 127310"},"PeriodicalIF":7.5,"publicationDate":"2025-03-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143724080","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Broad information diffusion modelling for sharing link click prediction using knowledge graphs","authors":"Xiangjie Kong ,&nbsp;Can Shu ,&nbsp;Lingyun Wang ,&nbsp;Hanlin Zhou ,&nbsp;Linan Zhu ,&nbsp;Jianxin Li","doi":"10.1016/j.eswa.2025.127276","DOIUrl":"10.1016/j.eswa.2025.127276","url":null,"abstract":"<div><div>In the new media era, users actively share and diffuse information across social networks, creating complex patterns of broad information diffusion (BID) that differ significantly from traditional recommendation scenarios. Existing models are primarily designed for deep information diffusion (DID) with sequential cascades and struggle to address BID challenges, including the sparse graph structure, weak temporal correlation, and ambiguity in user preferences. To bridge this gap, we propose K-BID, a knowledge-driven framework tailored for BID scenarios. K-BID integrates semantic and social graph information through a two-phase ‘Match &amp; Rank’ approach. The matching phase retrieves candidate voters using social relationships and personalized preferences, whereas the ranking phase refines predictions by modelling temporal dynamics. Experiments on real-world datasets demonstrate the superiority of K-BID over state-of-the-art methods, achieving significant improvements of 14.02%, 16.80%, and 16.99% in Precision, MRR, and AUC respectively, for the ‘Soc.’ objective with <span><math><mrow><mi>K</mi><mo>=</mo><mn>5</mn></mrow></math></span>. Our work advances the understanding of BID scenarios and offers a practical solution for optimizing information dissemination in social platforms.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"277 ","pages":"Article 127276"},"PeriodicalIF":7.5,"publicationDate":"2025-03-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143697236","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Unsupervised motion artifacts reduction for cone-beam CT via enhanced landmark detection","authors":"Thanaporn Viriyasaranon ,&nbsp;Serie Ma ,&nbsp;Mareike Thies ,&nbsp;Andreas Maier ,&nbsp;Jang-Hwan Choi","doi":"10.1016/j.eswa.2025.127258","DOIUrl":"10.1016/j.eswa.2025.127258","url":null,"abstract":"<div><div>Motion artifacts in cone-beam computed tomography (CBCT) primarily result from patient movement during the scanning process, which can compromise diagnostic accuracy. Emerging deep learning-based techniques have shown promise in mitigating these artifacts; however, they often rely on motion-free CBCT reconstructions for training, which poses practical challenges in clinical settings. An alternative approach involves leveraging the positions of metallic fiducial markers for motion estimation. While effective, this method is time-intensive and requires additional equipment installation, limiting its practicality. To address these challenges, we propose the Dynamic Landmark Motion Estimation (DLME) method, designed to reduce high-frequency noise and errors in landmark detection, thereby enhancing image quality. DLME is powered by the proposed TriForceNet, a novel landmark detection framework that integrates a sequential hybrid transformer-convolutional neural network architecture, multiresolution heatmap learning, and a multitask learning strategy augmented with an auxiliary segmentation head to improve motion estimation accuracy. Experimental evaluations demonstrate that TriForceNet achieves superior performance compared to state-of-the-art landmark detectors on two-dimensional projection images from the 4D extended cardiac-torso head phantom (XCAT) dataset, real patient CT scans from the CQ500 dataset, and knee regions from the CT scans in the VSD full body dataset. Furthermore, the DLME methodology outperforms traditional unsupervised motion compensation techniques and surpasses supervised, image-based motion artifact reduction methods across these datasets. The source code for the proposed model is publicly available at <span><span>https://github.com/Thanaporn09/TriForceNet.git</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"278 ","pages":"Article 127258"},"PeriodicalIF":7.5,"publicationDate":"2025-03-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143704058","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"CWFAS-Net: Low-light image enhancement using curvelet wavelet Attention and Fourier Transform","authors":"Hongfang Zhou ,&nbsp;Chenhui Cao ,&nbsp;Jiahao Tong ,&nbsp;Kangyun Zheng","doi":"10.1016/j.eswa.2025.127263","DOIUrl":"10.1016/j.eswa.2025.127263","url":null,"abstract":"<div><div>Significant advancements have been made in low-light image enhancement techniques, however, challenges remain regarding inconsistent restoration quality and unsatisfactory visual perception. To address these issues, we propose a robust and efficient method, CWFAS-Net. First, an Amplitude Illumination Estimation Module (AIEM) is constructed to enhance global brightness by amplifying amplitude components in the frequency domain. Second, the Curvelet-Wavelet Fourier Attention (CWFA) and Detail-Enhancement Attention (DEMA) modules are designed. CWFA combines features from the Fourier and wavelet domains to improve texture detail recovery and overall visual quality, while DEMA extracts local spatial features using a detail-enhancement attention mechanism. Finally, an SNR map is incorporated as a prior to guide information fusion within CWFA and DEMA. Using eight public benchmark datasets, both reference and non-reference metrics demonstrate that CWFAS-Net surpasses most mainstream algorithms, delivering superior enhancement and generalization in low-light restoration, thus validating our approach.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"278 ","pages":"Article 127263"},"PeriodicalIF":7.5,"publicationDate":"2025-03-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143734878","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Layered feature optimization framework based on Filtering, Embedding and data Balancing (L-FEB) for efficient DDoS attack detection","authors":"Rashmi Bhatia,&nbsp;Rohini Sharma","doi":"10.1016/j.eswa.2025.127230","DOIUrl":"10.1016/j.eswa.2025.127230","url":null,"abstract":"<div><div>The task of identifying Distributed Denial of Service (DDoS) attacks demands effective feature selection techniques to streamline the classification process without compromising accuracy. This research proposes a Layered framework based on Filtering, Embedding, and data Balancing (L-FEB) for feature optimization to enhance multi-classification in DDoS attack detection, an area that has been underrated in the literature. The L-FEB framework incorporates data preprocessing, feature selection through filtering and embedding techniques, and data balancing to reduce feature dimensionality while improving detection accuracy and efficiency. The study utilized five classification models: Random Forest (RF), XGBoost (XGB), k-Nearest Neighbors (k-NN), Multi-Layer Perceptron (MLP), and Long Short-Term Memory (LSTM) on the CICDDoS2019 dataset, which contains 13 types of DDoS attacks. The L-FEB framework reduced the feature set from 86 to 16, achieving 81.4% reduction. With the reduced feature set, all models demonstrated improved accuracy and reduced training and classification time. XGB achieved the highest accuracy of 85.8%, while MLP achieved the lowest training and classification time of 850 s. The MLP model was further optimized using a Triple-Layered Cross-Validation (TLCV) approach, reducing time by 75.37% while maintaining similar accuracy. The results demonstrate that the L-FEB framework effectively enhances both model performance and efficiency in multi-class DDoS attack detection.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"277 ","pages":"Article 127230"},"PeriodicalIF":7.5,"publicationDate":"2025-03-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143704837","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A novel method for identifying sudden degradation changes in remaining useful life prediction for bearing","authors":"Xianhua Chen,&nbsp;Zhigang Tian","doi":"10.1016/j.eswa.2025.127315","DOIUrl":"10.1016/j.eswa.2025.127315","url":null,"abstract":"<div><div>This paper presents a novel approach to enhancing the accuracy of Remaining Useful Life (RUL) predictions for bearings, addressing the limitations of traditional methods that fail to capture sudden changes in bearing health states. Conventional methods often rely on the monotonicity of a single feature, such as root mean square (RMS), and are unable to continuously monitor health changes. To overcome these challenges, a prototypical network is employed to identify sudden changes in bearing health states, referred to as critical points in this paper. By comparing the health states at the current time and the initial time, the critical point can be determined through the results of the prototypical network. Once the critical point is identified, the hyperparameters of the prototypical network are fixed and transformed to enable RUL prediction. Consequently, RUL can be predicted following the critical point. Furthermore, the prototypical network updates the initial time and continuously analyses the vibration signal to determine the next critical point. This process repeats until no further signals are input. Moreover, validation on two public datasets demonstrates the effectiveness of the proposed method in improving RUL prediction accuracy. Results indicate that the proposed method enhances model precision in bearing RUL prediction through critical point detection (CPD). Incorporating CPD offers a novel perspective for RUL prediction in industrial bearing applications.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"278 ","pages":"Article 127315"},"PeriodicalIF":7.5,"publicationDate":"2025-03-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143724876","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Hybrid Attentive Graph Neural Networks for time series gene expression clustering","authors":"Eleni Giovanoudi,&nbsp;Dimitrios Rafailidis","doi":"10.1016/j.eswa.2025.127136","DOIUrl":"10.1016/j.eswa.2025.127136","url":null,"abstract":"<div><div>Understanding gene expression patterns over time is crucial for unraveling the complexities of biological processes. An essential aspect of analyzing time series gene expression data involves the clustering of genes with similar responses, where the expression levels of genes are measured at specific time points. However, the main challenge in working with these time series data lies in uncovering concealed information embedded within their temporal structure. In this paper, we propose a novel Hybrid model that incorporates Attentive Graph Neural Networks, namely HAGNET, designed for clustering of time series gene expressions. In particular, HAGNET integrates both global patterns and temporal dependencies through a hybrid architecture of Graph Neural Networks (GNNs). Moreover, to account for the dynamic relationships of time series, HAGNET employs a custom temporal attention mechanism that filters each subsequent graph during the training process. The effectiveness of HAGNET is demonstrated on both synthetic and real-world biological datasets, outperforming several state-of-the-art methods. Finally, for reproduction purposes, we make the implementation of HAGNET publicly available at <span><span>https://github.com/egiovanoudi/HAGNET</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"277 ","pages":"Article 127136"},"PeriodicalIF":7.5,"publicationDate":"2025-03-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143681652","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Predicting concrete-encased column behavior under uniaxial load using adaptive neural fuzzy systems","authors":"Lamiaa K. Idriss ,&nbsp;Yasser A.S. Gamal","doi":"10.1016/j.eswa.2025.127354","DOIUrl":"10.1016/j.eswa.2025.127354","url":null,"abstract":"&lt;div&gt;&lt;div&gt;This research explores innovative approaches to assess concrete-encased steel (CES) composite columns, advancing beyond traditional finite element analysis limitations. The study lies in developing a rapid assessment methodology for CES columns under both axial and uniaxial loading conditions to enhanced building rigidity. By implementing advanced computational methods, the research aims to precisely predict CES behavior and structural stability, with particular emphasis on identifying and quantifying critical parameters that significantly influence column response. The investigation systematically examines various design factors and their interactions, enabling more informed decisions in structural engineering practice. This comprehensive approach addresses current knowledge gaps in CES column behavior, ultimately contributing to the development of more efficient, reliable, and optimized structural design guidelines for engineering applications. The study employs machine learning to analyze the uniaxial compressive load behavior of concrete-encased steel (CES) slender square composite columns, utilizing the Adaptive Neuro-Fuzzy Inference System (ANFIS) to optimize the Sugeno Fuzzy Inference System. This approach provides test information for calibrating Eurocode 4 against BS EN 1994-1, incorporating initial overall geometric imperfections (out-of-straightness) into the model. The finite element model, implemented using the software ABAQUS, has been validated against published experimental results for square cross-section (250*250 mm) CES columns with high-strength concrete (C100). The methodology encompasses 24 different cases, offering various predictions and test results for CES columns to ensure a comprehensive analysis of their structural behavior. The calculation of sensitivities for five different input parameters is considered: length of column (L), effective length (L&lt;sub&gt;eff&lt;/sub&gt;), slenderness ratio (SR), amount of compression load (P&lt;sub&gt;uniaxial&lt;/sub&gt;), and eccentricity of uniaxial load (e). The research reveals that to minimize vertical displacement (Z&lt;sub&gt;dis&lt;/sub&gt;) in concrete-encased steel (CES) columns, it is essential to avoid combinations with high values between length (L) or effective length (L&lt;sub&gt;eff&lt;/sub&gt;) and slenderness ratio (SR). Additionally, the study shows that the interaction between eccentricity (e) and uniaxial load(P&lt;sub&gt;uniaxial&lt;/sub&gt;) has a less pronounced effect on vertical displacement when compared to the significant impact observed when SR increases along with L&lt;sub&gt;eff&lt;/sub&gt; providing a theoretical basis for findings and their implications for column design and rapid assessment methodology for CES columns under axial and uniaxial loading conditions. This suggests that variations in P&lt;sub&gt;unaxial&lt;/sub&gt;, even when considered in conjunction with other parame-ters such as L, L&lt;sub&gt;eff&lt;/sub&gt;, and SR, exert only a modest influence on vertical displacement. Conversely, the combined effect of increasin","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"278 ","pages":"Article 127354"},"PeriodicalIF":7.5,"publicationDate":"2025-03-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143724564","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}