Expert Systems with Applications最新文献

{"title":"Phase-Glue: An interpretable and efficient backscattered electron-energy dispersive X-ray spectroscopy (BSE-EDS) mapping analysis approach to dissect the phase assemblage of cementitious material systems","authors":"Lihui Li ,&nbsp;Shuai Ding ,&nbsp;Yuxin Cai","doi":"10.1016/j.eswa.2025.127923","DOIUrl":"10.1016/j.eswa.2025.127923","url":null,"abstract":"<div><div>Backscattered electron and energy dispersive X-ray spectroscopy (BSE-EDS) mapping analysis has merged as a potent spatial-resolved technique in the research of cementitious materials, offering unprecedented opportunities in depicting complex phase assemblage and microstructure changes. Here, a data-driven and knowledge-informed BSE-EDS mapping analysis approach Phase-Glue is proposed, a visualized and versatile method designed to characterize phase assemblage, phase evolution, phase interaction in cementitious material systems. Phase-Glue ingeniously integrates BSE and EDS data, utilizing machine learning algorithms for automatic phase identification. It enables the correction of the identification results using expert knowledge and clarifies the correlations between the phases in embedding space, element features, and their actual spatial distributions through multi-view linked and shared analysis. The advantages of Phase-Glue are demonstrated through in-depth analysis of three cementitious systems (OPC, LC³, AAFS). Phase-Glue delineates the hydration of C₂S/C₃S, quantifies AFt/AFm and Hc/Mc, uncovers the component heterogeneity of fly ash, and distinguishes the different reaction products formed in AAFS system. Phase-Glue can characterize the multiple-modal features of phases in cementitious systems, providing insights into how phase alterations may underline the microstructure and properties changes of systems.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"283 ","pages":"Article 127923"},"PeriodicalIF":7.5,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143895695","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":"Fuzzy adaptive event-triggered frequency regulation for power system with hybrid energy storage under uncertain cyber-attacks","authors":"Lin Gan ,&nbsp;Kaibo Shi ,&nbsp;Jin Yang ,&nbsp;Hao Chen ,&nbsp;Zhenzhen Zhang ,&nbsp;Shouming Zhong","doi":"10.1016/j.eswa.2025.127867","DOIUrl":"10.1016/j.eswa.2025.127867","url":null,"abstract":"<div><div>This paper investigates the frequency regulation problem of power system with hybrid energy storage under uncertain cyber-attacks. Firstly, a switching model of multi-area interconnected power systems with hybrid energy storage under uncertain cyber-attacks is established. Then, a fuzzy adaptive event-triggered mechanism is proposed to reduce the data transmission. Based on this mechanism, a non-fragile controller is designed in the sense of control recovery. Further, the <span><math><msub><mrow><mi>H</mi></mrow><mrow><mi>∞</mi></mrow></msub></math></span> exponential stability criterion with less conservatism is established by proposing an improved piecewise asymmetric Lyapunov functional method. Finally, the particle swarm optimization algorithm is used to optimize the frequency regulation performance. Simulation examples are provided to validate the proposed strategies regarding the effectiveness in economic operation. The results demonstrate that the developed control strategy has intelligence to improve the power system operation stability with enhanced frequency regulation performance and less communication resources occupation. Meanwhile, total energy consumption is reduced in the presence of uncertain cyber-attacks.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"284 ","pages":"Article 127867"},"PeriodicalIF":7.5,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143903994","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 direct predictive DC-link voltage control with warm-starting iterative approach for three-level NPC BTB converter fed PMSG-based wind turbine systems","authors":"Mayilsamy Ganesh ,&nbsp;Jae Hoon Jeong","doi":"10.1016/j.eswa.2025.127747","DOIUrl":"10.1016/j.eswa.2025.127747","url":null,"abstract":"<div><div>This article presents a direct predictive DC-link voltage control (DPVC) scheme for a three-level (3L) neutral point clamped (NPC) back-to-back (BTB) converter fed permanent magnet synchronous generator (PMSG)-based wind turbine system (WTS). The conventional 3L NPC BTB converter employs an external PI controller to regulate the DC-link voltage at the set point by generating power or current references for the cascaded finite control set (FCS) predictive control. The presented strategy incorporates the DC-link voltage regulation term into the power regulation term in the grid-side converter (GSC) cost function, identifying the optimum switching state that directly regulates the DC-link voltage at the set point. Through this, the required active power is transferred to the grid to maintain power balance, and the conventional cascaded control structure is eliminated. In addition, a warm-starting iterative approach (WSIA) is introduced to derive the switching states employed for each switching instant. This reduces computation in both the GSC and machine-side converter (MSC) control and inherently incorporates switching frequency optimization. The presented scheme is validated in 3 MW rated WTS simulations under various operating conditions, with comparative studies performed to demonstrate its significance.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"283 ","pages":"Article 127747"},"PeriodicalIF":7.5,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143891868","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":"Speaker identification combining role knowledge graph correction and contextual block attention","authors":"Ye Tao ,&nbsp;Fei Wang ,&nbsp;Yanchang Cai ,&nbsp;Wei Li","doi":"10.1016/j.eswa.2025.127758","DOIUrl":"10.1016/j.eswa.2025.127758","url":null,"abstract":"<div><div>Character dialogue play a crucial role in novels, serving as a key element for understanding both the plot and character relationships. With advancements in artificial intelligence and natural language processing, dialogue speaker recognition has seen significant progress. In this paper, we integrate a character role knowledge graph with a self-training mechanism to perform the speaker recognition on Chinese novel quotations using large-scale novel corpora. Quotes in novels can generally be classified as explicit or implicit. Implicit quotes require identifying speakers from extensive contextual information, which poses challenges for existing models in handling long and detail-rich contexts. In addition, existing end-to-end speaker recognition methods are ineffective because they do not fully consider the relationship between context and quotes. In this paper, we first propose a Narrative Unit-based Context Selection (NUCS) algorithm for determining the context to which quotes belong. Secondly, a speaker recognition method based on Role Knowledge Graph Correction (RKGC) and Contextual Block Attention (CBA) is proposed. The proposed role knowledge graph correction algorithm improves speaker attribution by deeply analyzing role entities, their relationships, and relevant trigger words within quotations. It then refines candidate speaker probabilities obtained from the previous module. Additionally, the algorithm captures character relationships within the context using role mappings from existing novels. The CBA method introduces a block attention mechanism to capture character relationships within the context. It effectively computes the probability of each character being the speaker and determines the start and end indices of the most probable speaker segments. This allows the model to focus more precisely on speaker-related content, leading to more accurate speaker predictions. Experimental results demonstrate that our approach achieves competitive performance: on our self-constructed CNSI dataset, it attains 91.9% EM and 93.3% F1 scores. Although slightly lower than the SOTA SPC method (EM 92.3%, F1 93.6%), our approach demonstrates significant advantages in contextual reasoning capabilities, particularly for complex multi-character dialogue scenarios.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"284 ","pages":"Article 127758"},"PeriodicalIF":7.5,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143900425","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":"The parallel alternating direction method of multipliers: Optimal step-size or preconditioning matrix","authors":"Jing Chen ,&nbsp;Lianyuan Cheng ,&nbsp;Yang Yi ,&nbsp;Quanmin Zhu","doi":"10.1016/j.eswa.2025.127826","DOIUrl":"10.1016/j.eswa.2025.127826","url":null,"abstract":"<div><div>Alternating direction method of multipliers (ADMM) can decompose a complex problem into several smaller, more manageable subproblems, which can be solved independently. This is particularly useful for large-scale problems. However, ADMM has a slow convergence rate, especially compared to other optimization methods. In this paper, an alternating direction method of multipliers (ADMM) using two different parallel techniques is studied. First, the convergence properties of ADMM are given which can be regarded as instructions on how to design the modified ADMM. Then, by introducing the optimal step-size method and the preconditioning matrix method, the convergence rate can be increased, and researchers can use ADMM or its modifications to deal with different kinds of problems. Convergence analysis and numerical examples demonstrate our results.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"283 ","pages":"Article 127826"},"PeriodicalIF":7.5,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143886079","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":"AI-driven risk identification model for infrastructure project: Utilising past project data","authors":"Fredrick Ahenkora Boamah ,&nbsp;Xiaohua Jin,&nbsp;Sepani Senaratne,&nbsp;Srinath Perera","doi":"10.1016/j.eswa.2025.127891","DOIUrl":"10.1016/j.eswa.2025.127891","url":null,"abstract":"<div><div>Infrastructure projects are by nature complicated and vulnerable due to unpredictable risks encountered during their execution. The use of expert opinion and qualitative methodologies in traditional risk identification makes them vulnerable to subjectivity and responsiveness, leading to cost overruns, delays, and, ultimately, project failure. Therefore, to improve the accuracy of risk identification, this study utilises historical data in conjunction with AI approaches to develop a data-driven risk identification model. The model determines risk frequency and consequence by matching them to different risk categories in previous projects, considering word semantics. This model also demonstrates the facilitation of proactive decision-making and allows infrastructure project team members to identify risks early and implement mitigation plans. The study also highlights the practical significance of utilising historical data to make risk management strategies for infrastructure projects more reliable and efficient.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"283 ","pages":"Article 127891"},"PeriodicalIF":7.5,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143891866","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 Plug-in for cognitive diagnosis method based on correlation representation under long-tailed distribution","authors":"Yuhong Zhang ,&nbsp;Tiancheng He ,&nbsp;Shengyu Xu ,&nbsp;Mian Wang ,&nbsp;Chenyang Bu ,&nbsp;Xuegang Hu","doi":"10.1016/j.eswa.2025.127952","DOIUrl":"10.1016/j.eswa.2025.127952","url":null,"abstract":"<div><div>Cognitive diagnosis is a fundamental task in intelligence education, which aims to discover students’ proficiency for specific knowledge concepts. Existing cognitive diagnosis models are trained on the basis of sufficient student response records. In applications, however, these records usually follow a long-tailed distribution, i.e. there are only a few students with sufficient records, and a large number of students with a handful of records. The sparsity of records poses a challenge for cognitive diagnosis. To this end, a plug-in based on correlation representation is proposed to address cognitive diagnosis under long-tailed distribution, in which, the correlation representation between head students and tail students is learned to address the sparsity of long-tailed records. In particular, correlation representations are learned in view of both the cognitive state and the learning mode, which are learned based on the node representation and the subgraph representation, respectively. The correlation representation is then used as a plug-in to enhance the representation of long-tailed students and their related exercise and knowledge concepts. With the enhanced representations, the diagnostic performance of tail students is improved. Extensive experiments evaluate the improvement for diagnosis performance and the good compatibility of our plug-in component. Our code is available at <span><span>https://github.com/joyce99/Wangmian</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"284 ","pages":"Article 127952"},"PeriodicalIF":7.5,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143902190","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":"Y-Net-ECG: A Multi-Lead informed and interpretable architecture for ECG segmentation across diverse rhythms","authors":"Yuhang Liu ,&nbsp;Peng Zhang ,&nbsp;Xiaoli Feng ,&nbsp;Die Hu ,&nbsp;Ding Zhou ,&nbsp;Jingting Li ,&nbsp;Kaibiao Huang ,&nbsp;Yinuo Zhao ,&nbsp;Zuoming Fu ,&nbsp;Qianqian Zheng ,&nbsp;Zhigang Ye ,&nbsp;Tao Wang ,&nbsp;Xiaoyun Yang ,&nbsp;Fan Lin ,&nbsp;Qiang Li","doi":"10.1016/j.eswa.2025.127955","DOIUrl":"10.1016/j.eswa.2025.127955","url":null,"abstract":"<div><div>Accurate electrocardiogram (ECG) segmentation is critical for diagnosing and monitoring cardiac conditions. However, the accuracy of ECG segmentation across different heart rhythm types remains a challenge, and its practical utility in disease diagnosis remains to be fully validated. To address these challenges, we propose Y-Net, a deep learning model designed to perform robust ECG segmentation under both single-lead and multi-lead input modes. The model incorporates a dual-branch structure and a two-stage training strategy to ensure adaptability across various clinical scenarios. We evaluated Y-Net on two 12-lead ECG segmentation datasets: LUDB, a public dataset, and RDB, a privately annotated dataset based on public data but annotated specifically by our team. Y-Net demonstrated robust performance across datasets and rhythm types, achieving F1 scores of 99.60% and 99.44% in intra-dataset evaluations, and 99.03% and 98.24% in inter-dataset tests. To improve interpretability, we introduce an intermediate feature visualization method and apply segmentation results directly to atrial fibrillation (AF) detection based on P-wave absence. This morphology-based approach achieves AUCs of 0.946, 0.971, and 0.983 on the PhysioNet2017, CPSC2018, and AFDB datasets, respectively, without the need for additional classifiers. These results highlight the effectiveness and clinical potential of Y-Net as a transparent and adaptable tool for ECG segmentation and interpretation across diverse cardiac rhythms.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"283 ","pages":"Article 127955"},"PeriodicalIF":7.5,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143891865","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":"Profit scoring with machine learning in marketplace lending: Incorporating soft information","authors":"Jianwen Li ,&nbsp;Huicong Liang ,&nbsp;Yifei WanYan ,&nbsp;Lu Yu","doi":"10.1016/j.eswa.2025.127893","DOIUrl":"10.1016/j.eswa.2025.127893","url":null,"abstract":"<div><div>This paper examines the impact of non-semantic and semantic textual features on profit scoring in marketplace lending and compares the out-of-sample performance of machine learning models to that of traditional linear regression models. We find that the inclusion of soft information modestly enhances the model’s explanatory power, with certain aspects of soft information demonstrating significant predictive power for return rates. While linear machine learning models that employ regularization, such as Lasso, Bayesian Ridge, and Elastic Net, offer some advantages, none of them significantly outperform the baseline OLS model in out-of-sample performance. In contrast, most non-linear machine learning models, including the tree-based and instance-based models as well as feedforward neural network models, significantly outperform the OLS models, particularly when soft information is incorporated. Our findings shed lights on improving decision-making efficiency by incorporating soft information and employing non-linear machine learning models.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"284 ","pages":"Article 127893"},"PeriodicalIF":7.5,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143903274","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":"Log-driven predictive analysis of remaining time for emergency response processes","authors":"Rui Cao ,&nbsp;Qingtian Zeng ,&nbsp;Shuai Guo ,&nbsp;Weijian Ni ,&nbsp;Hua Duan ,&nbsp;Wenyan Guo","doi":"10.1016/j.eswa.2025.127800","DOIUrl":"10.1016/j.eswa.2025.127800","url":null,"abstract":"<div><div>The cross-organizational, messaging and resource attributes of emergency response processes effectively improve the accuracy of remaining time prediction. For this purpose, a log-driven analysis method for remaining time prediction of emergency response processes is proposed. Firstly, the attributes such as cross-organization, messaging and resources of the emergency response process are encoded, and then the vector representation of the emergency response process is obtained. Secondly, the vector representations of the emergency response processes are fed into a deep neural network prediction model for learning. Finally, we experimented with an emergency response process log to demonstrate the proposed method.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"283 ","pages":"Article 127800"},"PeriodicalIF":7.5,"publicationDate":"2025-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143882344","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}