Expert Systems with Applications最新文献

{"title":"A two-stage HV-driven adaptive multi-objective evolutionary algorithm and its application in fixed polarity reed-muller circuits","authors":"Lu Yang ,&nbsp;Shengsheng Wang ,&nbsp;Ruyi Dong ,&nbsp;Zihao Fu","doi":"10.1016/j.eswa.2025.128173","DOIUrl":"10.1016/j.eswa.2025.128173","url":null,"abstract":"<div><div>To achieve a balance between convergence and diversity, we proposed a two-stage HV-driven adaptive multi-objective evolutionary algorithm (TSAMEA). TSAMEA employs a sinusoidal decreasing parameter adjustment method to enhance exploration pace in the first stage. An adaptive parameter control mechanism utilizes historical memory pools and an HV-driven degree adjustment strategy to achieve better exploitation in the second stage. Extensive experimental data demonstrate that TSAMEA outperforms nine other compared MOEAs. The component analysis illustrates the efficacy of each component of TSAMEA. In addition, area and power optimization are now the main limitations in chip design, TSAMEA is applied to area and power optimization for Fixed Polarity Reed-Muller (FPRM) logic circuits and perform well, which further verifies the ability of the TSAMEA to solve practical problems.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"288 ","pages":"Article 128173"},"PeriodicalIF":7.5,"publicationDate":"2025-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144154951","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 spatial-temporal trend-event decoupling dual-channel framework for traffic flow prediction","authors":"Yuehai Xu,&nbsp;Lai Wei,&nbsp;Lu Feng","doi":"10.1016/j.eswa.2025.128107","DOIUrl":"10.1016/j.eswa.2025.128107","url":null,"abstract":"<div><div>Accurate traffic flow prediction is crucial for urban traffic control, route planning, and congestion detection. However, traffic data is influenced by spatial-temporal relationships and exhibits significant distribution drifts. This phenomenon can be attributed to volatile events in the traffic network, which make periodic trends ambiguous and difficult to learn. Consequently, traffic signals can be seen as a combination of fluctuating event signals and stable trend signals, both possessing rich and distinct spatial-temporal characteristics. Although recent methods have achieved considerable performance, most of them still roughly treat the traffic flow as a whole without considering the interactions between trend and event factors from a decoupled perspective. To address this issue, we propose a Spatial-Temporal Trend-Event Decoupling Dual-Channel Framework (TEDDCF) for traffic forecasting. TEDDCF first decomposes traffic flow into trend and event signals, and constructs a Dual-Channel Signal Encoder (DCSE) to model each signal independently. Temporally, DCSE uses multi-head attention and causal convolution to learn long-term trends and short-term event features. Spatially, we design two novel dynamic fusion graph convolutional modules-Trend-GCN and Event-GCN-to capture the independent spatial characteristics of each signal. In the decoder, the complete spatial-temporal representation of traffic flow is obtained through a Trend-Event Interactive Fusion (TEIF) module for prediction. Experiments on six traffic datasets show that TEDDCF outperforms state-of-the-art baseline models in prediction performance while significantly reducing computational costs. The source code is available at <span><span>https://github.com/XYHSMU/TEDDCF</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"287 ","pages":"Article 128107"},"PeriodicalIF":7.5,"publicationDate":"2025-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144090154","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":"Federal parameter-efficient fine-tuning for speech emotion recognition","authors":"Haijiao Chen ,&nbsp;Huan Zhao ,&nbsp;Zixing Zhang ,&nbsp;Keqin Li","doi":"10.1016/j.eswa.2025.128154","DOIUrl":"10.1016/j.eswa.2025.128154","url":null,"abstract":"<div><div>Pre-trained speech models leverage large-scale self-supervised learning to create general speech representations, with fine-tuning on specific tasks like Speech Emotion Recognition (SER) significantly enhancing performance. However, fine-tuning on different datasets necessitates storing full copies of model weights, leading to substantial storage demands and deployment challenges, particularly on resource-constrained devices. Centralized training also poses substantial privacy risks due to direct access to raw data. To address these challenges, we propose a cloud-edge-terminal collaborative paradigm for <u>Fed</u>eral <u>L</u>earning <u>P</u>arameter-<u>E</u>fficient <u>F</u>ine-<u>T</u>uning (FedLPEFT), which harnesses the synergy of cloud and edge computing to drive the development of collaborative SER applications. Specifically, the distributed paradigm of Federated Learning (FL) offers a privacy-preserving schema for collaborative training, and fine-tuning based on pre-trained speech models can improve SER performance. Parameter-Efficient Fine-Tuning (PEFT) embeds trainable layers in the feed-forward layers of pre-trained speech models. By freezing backbone parameters and sharing only a small set of trainable parameters, PEFT reduces communication overhead and enables lightweight interactions. Additionally, our experiments on attribute inference attacks across various pre-trained models show that gender prediction is at chance levels, indicating that the FedLPEFT approach significantly mitigates sensitive information leakage, ensuring robust privacy protection.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"287 ","pages":"Article 128154"},"PeriodicalIF":7.5,"publicationDate":"2025-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144090146","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":"Machine fault diagnosis method using Ghost-AdderNet and WSNs with sensor computing","authors":"Liqun Hou,&nbsp;Guopeng Mao,&nbsp;Ziming Zhang","doi":"10.1016/j.eswa.2025.128157","DOIUrl":"10.1016/j.eswa.2025.128157","url":null,"abstract":"<div><div>This paper proposes a machine fault diagnosis method using AdderNet with Ghost modules (Ghost-AdderNet) and wireless sensor networks (WSNs) with sensor computing. The proposed Ghost-AdderNet is a specially designed lightweight convolutional neural network (CNN) for machine fault diagnosis on resource-constrained WSN sensor nodes. It reduces the model size and computational cost by replacing the multiplication operations in the CNN with additions or subtractions while decreasing the model parameters by using Ghost modules. The proposed fault diagnosis method is verified by embedding and evaluating the designed Ghost-AdderNet on a commercial WSN node, JN5169 from NXP. The results show that, compared with raw data transmission mode, the proposed method can significantly reduce model size and the payload transmission data of WSNs, and save 25.1 mJ (29.1 %) node energy while maintaining acceptable diagnosis accuracy (above 99.6 %).</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"287 ","pages":"Article 128157"},"PeriodicalIF":7.5,"publicationDate":"2025-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144117133","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":"TL-ESKF: An information fusion method for INS/GPS integrated navigation considering driving state deviation","authors":"Yafei Wang ,&nbsp;Dongyu Luo ,&nbsp;Jiangfeng Wang ,&nbsp;Chongkai Qi ,&nbsp;Zhuan Chen ,&nbsp;Xuedong Yan","doi":"10.1016/j.eswa.2025.128168","DOIUrl":"10.1016/j.eswa.2025.128168","url":null,"abstract":"<div><div>When the global positioning system (GPS) signal is unavailable, the positioning performance of the GPS/inertial navigation system (INS) integrated navigation system significantly degrades, leading to deviation in the vehicle driving state. To address the limitations of information fusion during GPS outages and enhance navigation performance, this paper proposes an information fusion method based on transfer-learning error-state Kalman filter (TL-ESKF). This method consists of two steps, each combining a long short-term memory (LSTM) network with an ESKF. First, the dependence of the vehicle’s current position on historical INS and GPS data is considered, and the relationship between the gain of ESKF-1 and the observation vectors is established through the LSTM, providing an initial correction for the vehicle driving state deviation. Then, a second LSTM-ESKF combination further refines the correction, with transfer learning employed to expedite the training process of the TL-LSTM. Finally, the effectiveness of the proposed method is evaluated using both public datasets and real field tests. The experimental results indicate that, during GPS signal outages, the proposed method delivers more accurate navigation solutions. In comparison to high-precision machine learning methods, it offers a significantly higher training efficiency.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"287 ","pages":"Article 128168"},"PeriodicalIF":7.5,"publicationDate":"2025-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144117134","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 multi-scale spatial–temporal interaction fusion network for digital twin-based thermal error compensation in precision machine tools","authors":"Chi Ma ,&nbsp;Rongfeng Mu ,&nbsp;Mingming Li ,&nbsp;Jialong He ,&nbsp;Chunlei Hua ,&nbsp;Liang Wang ,&nbsp;Jialan Liu ,&nbsp;Giovanni Totis ,&nbsp;Jun Yang ,&nbsp;Kuo Liu ,&nbsp;Yuansheng Zhou ,&nbsp;Jianqiang Zhou ,&nbsp;Xiaolei Deng ,&nbsp;Shengbin Weng","doi":"10.1016/j.eswa.2025.127812","DOIUrl":"10.1016/j.eswa.2025.127812","url":null,"abstract":"<div><div>The machining accuracy of precision machine tools (PMTs) directly determines the quality of high-accuracy and complex components and thermal error (TE) significantly affects the machining accuracy of PMTs. The TE compensation is an effective way to reduce its effect and improve the machining accuracy. But the real-time performance of the TE compensation system and the prediction performance and robustness of the TE model are weak. In this study, a multi-scale spatial–temporal interaction fusion network (MSIFN) is designed and embedded into a digital twin framework for TE compensation to address the above issues. The efficient multi-scale squeeze-and-excitation network (EMSENet), spatial graph convolutional network (SGCN), and gated recurrent unit-temporal convolutional network (GRU-TCN) modules are designed for the MSIFN model to comprehensively capture and integrate spatial–temporal behaviors of thermal data. The EMSENet module is designed to emphasize critical features and suppress noise through multi-scale and channel attention mechanisms. The SGCN is able to realize accurate spatial relationship modeling, while the GRU-TCN is used to fuse spatial and temporal features, enhancing predictive accuracy and robustness. A lightweight digital twin-based TE compensation system is proposed, integrating the MSIFN model for real-time prediction and dynamic updates. Experimental results demonstrate the superior predictive performance of MSIFN, achieving a 38.9 % reduction in root mean square error and enhanced robustness compared to baseline models. Moreover, the total executing time of the TE compensation system based on the perception control-edge-cloud framework is reduced by 50.2 % compared with that of the TE compensation system based on the mist-cloud framework and that the reduction rates of the machining error is in the range of [61.5 %, 83.33 %] and [82.2 %, 83.3 %] at the initial and thermal states. This study provides a robust solution for improving machining accuracy in complex industrial environments.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"286 ","pages":"Article 127812"},"PeriodicalIF":7.5,"publicationDate":"2025-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144072104","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":"FL-Evo: Jointly modeling fact and logic evolution patterns for temporal knowledge graph reasoning","authors":"Ruishen Liu ,&nbsp;Xinzhi Wang ,&nbsp;Shaorong Xie ,&nbsp;Xiangfeng Luo ,&nbsp;Huizhe Su","doi":"10.1016/j.eswa.2025.128081","DOIUrl":"10.1016/j.eswa.2025.128081","url":null,"abstract":"<div><div>Temporal knowledge graphs (TKGs) extrapolation reasoning, intending to predict future events given the known KG sequence, benefits broad applications like policy-making and financial analysis. The key to this issue is to discern how knowledge evolves within these sequences. Currently, most works focus on modeling the evolution patterns through continuous sampling from TKGs, without ensuring the samples contain relevant facts or considering the knowledge beyond the samples. Faced with these challenges, we propose a novel model that performs prediction by capturing fact and logic knowledge evolution patterns (FL-Evo). For modeling fact evolution pattern, the fact knowledge is first distilled from large language models using designed prompts and subsequently refined with TKG. Then, entity-based subgraph sampling strategy extracts relevant facts from the TKG, capturing fact evolution patterns. Furthermore, logical knowledge mined from the TKG helps to derive the corresponding evolution pattern. Finally, the outputs of these two evolution patterns are integrated to realize the final prediction. Experimental results on five benchmark datasets demonstrate that FL-Evo outperforms existing temporal knowledge graph reasoning models, with improvements of up to 3.97 % in Hit@3 and 4.07 % in Hit@10. Notably, FL-Evo substantially enhances reasoning performance for unseen entities lacking prior records.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"286 ","pages":"Article 128081"},"PeriodicalIF":7.5,"publicationDate":"2025-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144068678","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 hybrid TwinSVM-HHO model for multilingual spam review detection using sentiment features and pre-trained embeddings","authors":"Ala’ M. Al-Zoubi ,&nbsp;Antonio M. Mora ,&nbsp;Hossam Faris ,&nbsp;Raneem Qaddoura","doi":"10.1016/j.eswa.2025.128160","DOIUrl":"10.1016/j.eswa.2025.128160","url":null,"abstract":"<div><div>The detection of spam reviews in multilingual environments remains a challenging task due to linguistic diversity, data imbalance, and semantic complexity. This paper proposes a novel hybrid model that integrates Twin Support Vector Machine (TwinSVM) with Harris Hawks Optimization (HHO) for simultaneous parameter optimization and feature selection. To enhance semantic understanding, sentiment-based features are incorporated alongside pre-trained word embedding models—BERT, FastText, and MUSE—across English, Arabic, and Spanish datasets. Our approach generates 24 high-quality datasets using embeddings with 100 and 400 dimensions, including a combined multilingual set. Experimental results demonstrate that our proposed HHO-TwinSVM model consistently outperforms conventional classifiers and metaheuristic-enhanced SVMs, achieving accuracy improvements of up to 9.44 % and enhanced robustness in low-resource languages. This integrated framework represents a scalable and adaptable solution for multilingual spam detection. Four detailed experiments were conducted in this study, each designed to address and demonstrate a specific aspect of the proposed approach. Across all experiments, the method outperformed existing algorithms, achieving impressive accuracy rates of 92.9741 %, 89.0314 %, 80.3580 %, and 85.0859 % on Arabic, English, Spanish, and multilingual datasets, respectively. Subsequently, sentiment analysis features were incorporated to further enhance detection performance, resulting in improvements of 1.0994 %, 2.6674 %, 9.4430 %, and 8.7448 %, respectively. A comprehensive analysis of the experimental results, including the influence of reviews and sentiment features, is also presented.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"287 ","pages":"Article 128160"},"PeriodicalIF":7.5,"publicationDate":"2025-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144117131","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":"Soil moisture retrieval and trend prediction using multi-temporal remote sensing data: An interpretable deep regression approach","authors":"Xiaofei Kuang,&nbsp;Shiyu Xiang,&nbsp;Jiao Guo","doi":"10.1016/j.eswa.2025.128172","DOIUrl":"10.1016/j.eswa.2025.128172","url":null,"abstract":"<div><div>High-precision retrieval of soil moisture (SM) and prediction of its trends are crucial for research in agriculture, meteorology, and related fields. Multi-temporal multi-source remote sensing data can provide temporal variation information of various features. This study aims to achieve high-precision retrieval and prediction of SM by leveraging multi-temporal features. Compared to physics-based models, data-driven regression models exhibit significant advantages in handling multi-dimensional complex features. However, the lack of effective interpretation of their operational mechanisms remains a limitation in current data-driven model research. Given the superior performance of Transformer networks in processing multi-sequence features, this study constructs a deep regression model based on the Transformer architecture for SM extraction. To interpret the SM regression process of this model, the study quantifies the influence of input features on regression, analyzes the temporal variations of intermediate hidden features, and evaluates the output performance to elucidate the feature extraction and regression mechanisms. Experiments were conducted in the Pacific Northwest region. Analysis of feature derivatives and intermediate hidden features reveals that the Transformer intelligently allocates appropriate attention to data at different time points, resulting in stronger feature influence closer to the retrieval or prediction date. The experimental results indicate that multi temporal information is beneficial for SM retrieval and prediction, while assigning appropriate attention to features at different time points is more advantageous for predicting SM trends. This study provides a practical approach for deep regression-based SM retrieval and prediction and offers insights into interpreting the SM regression mechanisms of the Transformer.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"287 ","pages":"Article 128172"},"PeriodicalIF":7.5,"publicationDate":"2025-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144084266","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":"Boundary-aware dynamic re-weighting for semi-supervised medial image segmentation","authors":"Weili Jiang ,&nbsp;Jiawen Li ,&nbsp;Yihao Li ,&nbsp;Xifei Wei ,&nbsp;Jianping Huang ,&nbsp;Gwenolé Quellec ,&nbsp;Weixin Si ,&nbsp;Chubin Ou","doi":"10.1016/j.eswa.2025.128175","DOIUrl":"10.1016/j.eswa.2025.128175","url":null,"abstract":"<div><div>Different from traditional semi-supervised learning (SSL), semi-supervised medical image segmentation faces two significant challenges: (1) the imbalanced distribution of labeled data causes models to bias towards majority classes; (2) the distribution discrepancy between labeled and unlabeled samples induces confirmation bias in pseudo-labels. Inspired by clinical practice, where experienced doctors utilize intrinsic features from the interior of target organs to clarify ambiguous boundaries and focus on minority classes, we propose a novel boundary-aware dynamic re-weighting network (BDRN). First, we utilize edge filters to generate visually different but semantically aligned views, compelling two sub-networks to learn informative features from organ interiors and boundaries, respectively. Second, we extract the boundary and interior regions using morphological operators and introduce a shape constraint to enhance feature learning. Additionally, a conflict-adversarial module promotes segmentation consistency between different views. Finally, we propose a dynamic re-weighting strategy based on the effective number to improve attention to imbalanced classes. Experiments demonstrate that our method significantly improves segmentation performance, achieving state-of-the-art results on CT and MR images. Ablation studies further confirm the efficacy of boundary consistency constraints and dynamic re-weighting. The segmentation Dice score for minority organs (e.g., esophagus) on the Synapse dataset is improved by 17.1 %, 46.2 %, and 49.4 % using 10 %, 20 %, and 40 % labeled data, respectively.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"287 ","pages":"Article 128175"},"PeriodicalIF":7.5,"publicationDate":"2025-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144117129","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}