Expert Systems with Applications最新文献

{"title":"Data-driven algorithm for temperature predictions and corrections from low-resolution thermal images at fire scenes","authors":"Yichuan Dong,&nbsp;Jian Jiang,&nbsp;Wei Chen,&nbsp;Jihong Ye","doi":"10.1016/j.eswa.2025.127771","DOIUrl":"10.1016/j.eswa.2025.127771","url":null,"abstract":"<div><div>An accurate and efficient temperature measurement at fire scenes is crucial for structural safety predictions and fire emergency responses. The application of thermal images provides advantages of spatial and stable measurements over thermocouples. A data-driven algorithmic system for temperature measurement is proposed, utilizing thermal images and comprising a sequence of resolution enhancements, temperature predictions, and error corrections. The system starts with transformation of low-resolution images to super-resolution ones through convolutional neural networks (CNN) with hybrid scaling factors and attention fusion post-residual blocks. The temperatures are predicted from super-resolution thermal images based on cascade feedforward neural networks (CFNN) using a two-stage temperature division strategy. The errors of temperature predictions are corrected by comparing results between thermal images and thermocouples. The effectiveness, influencing factor and optimization strategy of the proposed system are validated through a series of large-scale fire tests. The mean absolute errors of temperature prediction models are within 20 °C, while over 70 % of error correction results are within ±30 °C. The proposed algorithm provides an effective tool to predict and correct temperature fields, aiming at a fast and smart fire emergency decision-making.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"282 ","pages":"Article 127771"},"PeriodicalIF":7.5,"publicationDate":"2025-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143855296","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":"Fact retrieval from knowledge graphs through semantic and contextual attention","authors":"Akhil Chaudhary ,&nbsp;Enayat Rajabi ,&nbsp;Somayeh Kafaie ,&nbsp;Evangelos Milios","doi":"10.1016/j.eswa.2025.127612","DOIUrl":"10.1016/j.eswa.2025.127612","url":null,"abstract":"<div><div>Knowledge Graphs (KGs), such as DBpedia and ConceptNet, enhance Natural Language Processing (NLP) applications by providing structured information. However, extracting accurate data from KGs is challenging due to issues in entity detection, disambiguation, and relation classification, which often lead to errors and inefficiencies. We introduce <strong>Attention2Query (A2Q)</strong>, an attention-driven approach that directly ranks and selects the most relevant facts, thus minimizing error propagation. A2Q centres on three key contributions: (1) <em>Focused Node Selection</em>, which streamlines graph traversal; (2) <em>Global Attention Alignment</em>, improving retrieval by comparing facts against the query text; and (3) <em>Contextual Re-ranking</em>, enabling on-the-fly adjustments of fact importance based on evolving query context. Experimental results across multiple tasks and datasets show that A2Q substantially outperforms baseline methods, including those in zero-shot settings, achieving higher retrieval accuracy with reduced computational overhead.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"282 ","pages":"Article 127612"},"PeriodicalIF":7.5,"publicationDate":"2025-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143869515","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":"SABA: Scale-adaptive Attention and Boundary Aware Network for real-time semantic segmentation","authors":"Huilan Luo ,&nbsp;Chunyan Liu ,&nbsp;Lik-Kwan Shark","doi":"10.1016/j.eswa.2025.127680","DOIUrl":"10.1016/j.eswa.2025.127680","url":null,"abstract":"<div><div>Balancing accuracy and speed is crucial for semantic segmentation in autonomous driving. While various mechanisms have been explored to enhance segmentation accuracy in lightweight deep learning networks, adding more mechanisms does not always lead to better performance and often significantly increases processing time. This paper investigates a more effective and efficient integration of three key mechanisms — context, attention, and boundary — to improve real-time semantic segmentation of road scene images. Based on an analysis of recent fully convolutional encoder–decoder networks, we propose a novel Scale-adaptive Attention and Boundary Aware (SABA) segmentation network. SABA enhances context through a new pyramid structure with multi-scale residual learning, refines attention via scale-adaptive spatial relationships, and improves boundary delineation using progressive refinement with a dedicated loss function and learnable weights. Evaluations on the Cityscapes benchmark show that SABA outperforms current real-time semantic segmentation networks, achieving a mean intersection over union (mIoU) of up to 76.7% and improving accuracy for 17 out of 19 object classes. Moreover, it achieves this accuracy at an inference speed of up to 83.4 frames per second, significantly exceeding real-time video frame rates. The code is available at <span><span>https://github.com/liuchunyan66/SABA</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"282 ","pages":"Article 127680"},"PeriodicalIF":7.5,"publicationDate":"2025-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143864810","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":"Learning adaptive detection and tracking collaborations with augmented UAV synthesis for accurate anti-UAV system","authors":"Shihan Liu ,&nbsp;Tianyang Xu ,&nbsp;Xue-Feng Zhu ,&nbsp;Xiao-Jun Wu ,&nbsp;Josef Kittler","doi":"10.1016/j.eswa.2025.127679","DOIUrl":"10.1016/j.eswa.2025.127679","url":null,"abstract":"<div><div>In recent years, Unmanned Aerial Vehicles (UAVs) have witnessed a significant use upsurge across various domains. However, their low cost and elusive size have raised significant security concerns. In particular, UAVs, imaged in the infrared mode, located in a cluttered background and being relatively small targets, pose formidable challenges. To this end, in this paper, we propose a novel approach for accurate UAVs perception. The core innovation lies in learning an adaptive detection and tracking collaboration mechanism, supported by a novel method of training data augmentation (ADTC). During the test phase, ADTC begins by leveraging the detector to identify the potential target candidates within the image frame. These candidates are then refined by an adaptive selection module, where a Kalman filter is deployed to model and predict the motion trajectory of each candidate. The best result of the predicted trajectories with the detected candidates is adopted as the output. The adaptive selection module filters out less confident objects, efficiently decreasing processing time. Furthermore, we construct a new dataset Anti-MUAV15 to evaluate the performance of ADTC in multiple-UAV scenarios. Our approach has been rigorously evaluated through qualitative and quantitative experiments on the Anti-UAV, AntiUAV600 and Anti-MUAV15 datasets. The experimental results demonstrate that our method outperforms state-of-the-art anti-UAV solutions in terms of robustness and precision, without imposing additional computational burden. The code and dataset are available at <span><span>https://github.com/Shihan0325/Anti-MUAV15</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"282 ","pages":"Article 127679"},"PeriodicalIF":7.5,"publicationDate":"2025-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143859341","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":"Real time brain stroke identification using face images based on machine learning and booby bird optimization","authors":"Alaa M. Mohamed ,&nbsp;Asmaa H. Rabie ,&nbsp;Hanan M. Amer ,&nbsp;Ahmed I. Saleh ,&nbsp;Mohy Eldin A. Abo-Elsoud","doi":"10.1016/j.eswa.2025.127719","DOIUrl":"10.1016/j.eswa.2025.127719","url":null,"abstract":"<div><div>Stroke one of the most common causes of death after heart disease and cancer and is the leading cause of severe long-term disability. The earlier a stroke is detected and the faster it is treated, the greater the chance of recovery. Therefore, early detection and treatment of strokes are essential to save lives and reduce permanent damage. In this paper, we present a stroke monitoring strategy based on face images for individuals who have a risk of stroke or those with chronic diseases. Patients are monitored using a smart camera located in their room. This camera sends data to a fog server in the hospital, where all processes are done there. The proposed strategy consists of three main stages, which are i) data preprocessing, ii) model training, and iii) model testing. In the data preprocessing stage, faces are extracted from images using Yolo v8. Then features are extracted using the Active Appearance Model (AAM) model and dlib library. The proposed Binary Booby Bird Optimization (B³O) is used as a new feature selection method to select only relevant features from data sets. In the model training stage, each feature is divided into ranges and these ranges are divided into regions to minimize the data set size for fast detection. Finally, the model testing stage tests all proposed stages to detect stroke patients using the Navie Bayes (NB) classifier. The experiment results show that the proposed B³O and proposed stroke monitoring strategy achieve high accuracy of 94.18% and 98.43%, respectively.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"282 ","pages":"Article 127719"},"PeriodicalIF":7.5,"publicationDate":"2025-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143851359","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":"CRC-MRC: Reader Comments Augmented Machine Reading Comprehension for social emotion prediction","authors":"Hengxin Gao ,&nbsp;Keyang Ding ,&nbsp;Qianlong Wang ,&nbsp;Bin Liang ,&nbsp;Ruifeng Xu","doi":"10.1016/j.eswa.2025.127336","DOIUrl":"10.1016/j.eswa.2025.127336","url":null,"abstract":"<div><div>The task of social emotion prediction aims to understand and predict the distribution of emotion that a text evokes in its readers. Previous research has primarily focused on modeling the textual representation of news while neglecting the human aspect of how news is read and the emotions it evokes. Thus, we utilize the Machine Reading Comprehension (MRC) framework to read articles like humans do. Additionally, previous studies have shown the significant help of integrating readers’ comments. However, in realistic scenarios, raw comments are not readily available and are often redundant and noisy. To address this, we suggest a clustering-based approach that utilizes LLMs for the automatic generation of comments, aiming to provide high-quality emotional information from the reader’s perspective. By integrating generated comments into the MRC framework, we propose a Clustering-based Reader Comments Augmented Machine Reading Comprehension framework (CRC-MRC) to comprehensively model the reading process from the readers’ perspective while browsing news and comments. Extensive tests on benchmark datasets demonstrate the high effectiveness of our proposed framework, surpassing current state-of-the-art methods.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"282 ","pages":"Article 127336"},"PeriodicalIF":7.5,"publicationDate":"2025-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143874000","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":"Diffulex: Diffusion based lexically constrained text generation with mixed absorbing state and constraint balance","authors":"Fengrui Kang,&nbsp;Xianying Huang,&nbsp;Bingyu Li","doi":"10.1016/j.eswa.2025.127614","DOIUrl":"10.1016/j.eswa.2025.127614","url":null,"abstract":"<div><div>Lexically constrained text generation aims to generate complete text with given keywords, which can be applied in many fields, such as dialogue systems, automatic summarization, and story generation. However, the current methods often find it difficult to strike a balance between generation quality, constraint ability, and generation speed, and most of them can only focus on one aspect, which seriously limits their applications. To solve this problem, we propose Diffulex, a lexically constrained text generation model based on the diffusion model, which achieves faster generation speed and higher flexibility. In response to the characteristics of lexically constrained text generation tasks, Diffulex employs a forward process of mixed absorbing states, converting tokens into different types of [MASK] tags to capture the semantic relationship between constraints and tokens. Through the constraint balance of the reverse process, more attention will be paid to the prediction tokens that meet the constraint conditions and promote the dynamic fusion of the hidden state constraint information, achieving the balance between the generation quality and the constraint ability. We compared Diffulex with advanced work in the field and popular large language models as baselines, and our results on multiple datasets show that Diffulex outperforms the baseline in various aspects. Our code is available on <span><span>https://github.com/Kenfree0/Diffulex</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"281 ","pages":"Article 127614"},"PeriodicalIF":7.5,"publicationDate":"2025-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143848627","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":"IoT-enabled recurrent spatio-temporal adaptive attention of temporal convolutional transformer with continual learning for dairy farming","authors":"G. Anand Kumar ,&nbsp;B. Muni Lavanya ,&nbsp;Md. Khaja Mohiddin ,&nbsp;Sourabh Mitra","doi":"10.1016/j.eswa.2025.127712","DOIUrl":"10.1016/j.eswa.2025.127712","url":null,"abstract":"<div><div>Cattle health and behavior monitoring is critical in the maintenance of livestock welfare and efficient farm productivity. However, meaningful features for a health alert system are hard to extract from voluminous data generated by IoT sensors, which track various health and environmental parameters. In this paper, a new system along these lines for challenges as described above is presented: an advanced alert system using the Recurrent Spatio-Temporal Adaptive Attention of Temporal Convolutional Transformer (RecSTAA-TCT) model. The newly proposed model has integrated the following major components: a dynamic residual bidirectional gated recurrent unit, intensive spatial attention, and a Temporal Adaptive Temporal Convolutional Transformer module. This makes feature extraction with time series data rather challenging due to the complexity and variability of the data generated by IoT sensors. The Adaptive Residual Bi-GRU achieves this by efficiently handling the temporal dependencies to improve the robustness of the model in the presence of missing data and noise. Critical spatial features are then extracted by intensive spatial attention, hence allowing the system to focus attention only on the most informative data in a biometric and environmental database. The Temporal Adaptive TCT module refines further the model capability by extracting temporal features to make precise predictions that trigger the alert system in response to possible health and behavioral anomalies. It embeds continual learning into the model, through which it learns new patterns and data with time, giving back predictive accuracy and reliability to the model. It is based on this foundation that an integrated approach provides proactive management and timely interventions, hence substantially improving real-time anomaly detection over the traditional methods of monitoring. The proposed RecSTAA-TCT model is giving a classification accuracy of 96.5% and delivering alert notifications at a response time of 10.3 s.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"282 ","pages":"Article 127712"},"PeriodicalIF":7.5,"publicationDate":"2025-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143859161","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":"LGSMOTE-IDS: Line Graph based Weighted-Distance SMOTE for imbalanced network traffic detection","authors":"Guyu Zhao,&nbsp;Linwei Li,&nbsp;Hongdou He,&nbsp;Jiadong Ren","doi":"10.1016/j.eswa.2025.127645","DOIUrl":"10.1016/j.eswa.2025.127645","url":null,"abstract":"<div><div>The application of Graph Neural Networks (GNNs) to Network Intrusion Detection Systems (NIDS) has become a prominent research focus. However, NIDS often struggles to classify minority attack samples due to the severe class imbalance in NIDS datasets, where the number of samples varies significantly across classes. Additionally, prior studies have frequently overlooked the importance of edge features in GNNs. To address these challenges, we propose LGSMOTE-IDS, a novel framework that integrates a <strong><u>L</u></strong>ine <strong><u>G</u></strong>raph based Weighted-Distance <strong><u>SMOTE</u></strong> for <strong><u>I</u></strong>ntrusion <strong><u>D</u></strong>etection <strong><u>S</u></strong>ystems. First, we define the fine-grained protocol service graph (<span><math><mrow><mi>P</mi><mi>S</mi><mi>G</mi></mrow></math></span>) and transform it into its corresponding protocol service line graph (<span><math><mrow><mi>L</mi><mrow><mo>(</mo><mi>P</mi><mi>S</mi><mi>G</mi><mo>)</mo></mrow></mrow></math></span>). This transformation provides a novel perspective for describing network traffic interactions and enables the conversion of the edge classification task into a node classification task. Second, we introduce Weighted-Distance SMOTE, an oversampling algorithm specifically tailored to NIDS datasets, which employs an improved interpolation strategy to generate synthetic minority class samples. Finally, we utilize a GNN-based classifier to predict labels for all samples. We conduct experiments on three widely used datasets—NF-UNSW-NB15, NF-BoT-IoT, and NF-ToN-IoT. LGSMOTE-IDS achieves average increases of 18.11%, 45.91%, and 36.41% in weighted F1-scores for five, one, and three minority classes across the three datasets, respectively, compared to baseline method. Moreover, LGSMOTE-IDS successfully detects attack types that previous models fail to recognize. To the best of our knowledge, LGSMOTE-IDS is the first framework to integrate GNNs with an oversampling algorithm to address the class imbalance issue in NIDS datasets.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"281 ","pages":"Article 127645"},"PeriodicalIF":7.5,"publicationDate":"2025-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143838609","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":"ShuiAttNet: Fully convolutional attention network for Shuishu character recognition","authors":"Xiaojun Bi ,&nbsp;Lu Han ,&nbsp;Weizheng Qiao","doi":"10.1016/j.eswa.2025.127613","DOIUrl":"10.1016/j.eswa.2025.127613","url":null,"abstract":"<div><div>Shuishu is one of the most representative hieroglyphs and precious cultural heritage in China, currently facing the risk of extinction. Preserving this endangered script requires innovative approaches to accurately recognize its characters. However, existing methods face significant challenges, including the inability to handle the broad diversity of Shuishu characters and the complexities of authentic ancient manuscripts. To address these issues, we present a comprehensive study that combines dataset construction and advanced deep learning methods. First, we establish the largest and most diverse Shuishu single-character dataset named S842 to date, addressing the critical lack of publicly available resources for Shuishu. Then we propose a novel Fully Convolutional Attention Network named ShuiAttNet, which is specifically designed for Shuishu character recognition. ShuiAttNet introduces two key innovations: the Attentional MBConv (AMC) block and the Fully Convolutional Attention (FCA) block. The AMC block utilizes a novel feature fusion mechanism to capture fine-grained local details while reducing feature redundancy caused by the low-rank characteristics of Shuishu characters. Meanwhile, the FCA block employs Depthwise Separable Dilated Convolution to establish long-range dependencies while preserving the two-dimensional spatial structure of the images. These components enable ShuiAttNet to achieve superior performance with significantly fewer parameters compared to existing methods. Extensive experiments validate the effectiveness and superiority of ShuiAttNet in both quantitative and qualitative assessments. Experimental results show that our proposed model achieves a Top-1 Acc of 97.04%, outperforming other state-of-the-art methods.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"282 ","pages":"Article 127613"},"PeriodicalIF":7.5,"publicationDate":"2025-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143876889","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}