Expert Systems with Applications最新文献_第5页

A dynamic interactive consensus deviation correction system driven by hybrid intelligence and its application to NEV policies

IF 7.5 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Expert Systems with Applications

Pub Date : 2025-02-19 DOI: 10.1016/j.eswa.2025.126877

Jinpeng Wei , Xuanhua Xu , Weiwei Zhang , Qiuhan Wang

Countries worldwide continue to introduce policies to accelerate the development of the new energy vehicle (NEV) industry. However, during implementation, NEV policies often face target deviations due to complex real-world circumstances, leading to limited adoption and unintended negative consequences. While human intelligence has struggled to effectively assess and correct these deviations, machine reasoning and judgment offer significant potential to complement human intelligence. Despite this, previous research on NEV policies has inadequately explored the evaluation and correction of policy deviations, particularly via the use of emerging machine intelligence from artificial intelligence (AI) and big data. Studies suggest that machines can contribute to the challenging task of identifying NEV policy deviations through informed inference and feedback. This work proposes the integration of human and machine intelligence to address these deviations. Specifically, a dynamic interactive consensus system driven by hybrid intelligence is designed to correct NEV policy deviations. This system, based on the consensus-reaching process (CRP), uses a two-stage algorithm that includes deviation identification metrics and correction feedback iteration. Additionally, it leverages machine learning (ML) to extract hybrid intelligence judgments from online comments and text data, providing necessary information and parameters for deviation correction. An experimental analysis demonstrates the effectiveness of the proposed system and highlights the complementary benefits of combining human and machine intelligence.

{"title":"A dynamic interactive consensus deviation correction system driven by hybrid intelligence and its application to NEV policies","authors":"Jinpeng Wei , Xuanhua Xu , Weiwei Zhang , Qiuhan Wang","doi":"10.1016/j.eswa.2025.126877","DOIUrl":"10.1016/j.eswa.2025.126877","url":null,"abstract":"<div><div>Countries worldwide continue to introduce policies to accelerate the development of the new energy vehicle (NEV) industry. However, during implementation, NEV policies often face target deviations due to complex real-world circumstances, leading to limited adoption and unintended negative consequences. While human intelligence has struggled to effectively assess and correct these deviations, machine reasoning and judgment offer significant potential to complement human intelligence. Despite this, previous research on NEV policies has inadequately explored the evaluation and correction of policy deviations, particularly via the use of emerging machine intelligence from artificial intelligence (AI) and big data. Studies suggest that machines can contribute to the challenging task of identifying NEV policy deviations through informed inference and feedback. This work proposes the integration of human and machine intelligence to address these deviations. Specifically, a dynamic interactive consensus system driven by hybrid intelligence is designed to correct NEV policy deviations. This system, based on the consensus-reaching process (CRP), uses a two-stage algorithm that includes deviation identification metrics and correction feedback iteration. Additionally, it leverages machine learning (ML) to extract hybrid intelligence judgments from online comments and text data, providing necessary information and parameters for deviation correction. An experimental analysis demonstrates the effectiveness of the proposed system and highlights the complementary benefits of combining human and machine intelligence.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"273 ","pages":"Article 126877"},"PeriodicalIF":7.5,"publicationDate":"2025-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143464058","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}

引用次数: 0

Greedy-assisted teaching-learning-based optimization algorithm for cost-based hybrid flow shop scheduling

IF 7.5 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Expert Systems with Applications

Pub Date : 2025-02-19 DOI: 10.1016/j.eswa.2025.126955

Wasif Ullah , Mohd Fadzil Faisae Ab Rashid , Muhammad Ammar Nik Mu’tasim

Production scheduling is a strategic process that organizes the execution of jobs on available resources to optimize specific objectives. One significant scheduling challenge is the Cost-based Hybrid Flow Shop (CHFS) problem, which involves optimizing job scheduling across multiple stages to minimize scheduling-related costs. However, limited attention has been given to CHFS when considering holistic cost models using efficient algorithms. This paper presents a novel Greedy-Assisted Teaching-Learning-Based Optimization (GTLBO) algorithm for CHFS. Unlike previous studies that focus on isolated cost factors, this research formulated an integrated mathematical model for CHF holistically capturing labor, energy consumption, maintenance, and late penalty costs. The GTLBO algorithm incorporates a unique hybrid initialization strategy, generating 10 % of the initial population using a Greedy algorithm to enhance exploration efficiency. The performance of GTLBO was evaluated through computational experiments involving 12 test instances, with comparative algorithms included for analysis. Results from the Wilcoxon rank-sum test indicated a significant difference between the outputs of GTLBO and other algorithms, with GTLBO outperforming the comparative algorithms in 75 % of the test instances. Additionally, the case study validation showed that GTLBO can reduce costs by 0.23 % to 4.31 % compared to other algorithms. This research offers valuable insights for manufacturers seeking to optimize CHFS scheduling to reduce production expenses.

{"title":"Greedy-assisted teaching-learning-based optimization algorithm for cost-based hybrid flow shop scheduling","authors":"Wasif Ullah , Mohd Fadzil Faisae Ab Rashid , Muhammad Ammar Nik Mu’tasim","doi":"10.1016/j.eswa.2025.126955","DOIUrl":"10.1016/j.eswa.2025.126955","url":null,"abstract":"<div><div>Production scheduling is a strategic process that organizes the execution of jobs on available resources to optimize specific objectives. One significant scheduling challenge is the Cost-based Hybrid Flow Shop (CHFS) problem, which involves optimizing job scheduling across multiple stages to minimize scheduling-related costs. However, limited attention has been given to CHFS when considering holistic cost models using efficient algorithms. This paper presents a novel Greedy-Assisted Teaching-Learning-Based Optimization (GTLBO) algorithm for CHFS. Unlike previous studies that focus on isolated cost factors, this research formulated an integrated mathematical model for CHF holistically capturing labor, energy consumption, maintenance, and late penalty costs. The GTLBO algorithm incorporates a unique hybrid initialization strategy, generating 10 % of the initial population using a Greedy algorithm to enhance exploration efficiency. The performance of GTLBO was evaluated through computational experiments involving 12 test instances, with comparative algorithms included for analysis. Results from the Wilcoxon rank-sum test indicated a significant difference between the outputs of GTLBO and other algorithms, with GTLBO outperforming the comparative algorithms in 75 % of the test instances. Additionally, the case study validation showed that GTLBO can reduce costs by 0.23 % to 4.31 % compared to other algorithms. This research offers valuable insights for manufacturers seeking to optimize CHFS scheduling to reduce production expenses.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"273 ","pages":"Article 126955"},"PeriodicalIF":7.5,"publicationDate":"2025-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143464059","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}

引用次数: 0

Generalized reinforcement learning control algorithm for fully automated insulin delivery system

IF 7.5 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Expert Systems with Applications

Pub Date : 2025-02-19 DOI: 10.1016/j.eswa.2025.126909

Vega Pradana Rachim , Junyoung Yoo , Jaeyeon Lee , Yein Lee , Sung-Min Park

A fully automated insulin delivery (Fully-AID) system is expected to provide the ultimate safety, comfort, and a sense of freedom for people living with diabetes (PwD). Previous studies have shown the potential of a deep reinforcement learning (DRL) model for fully-AID control algorithm in simulation environment. However, the practical implementation is still challenging due to the domain gaps between simulation and real world scenario. In this manuscript, we proposed a novel generalized control algorithm, called xgDRL, to realize a DRL-driven fully-AID system. The generalization of the proposed algorithm is achieved by our two main contributions that are introducing a novel concept of fully-AID context called total daily insulin (TDI) into the input of DRL model, and a novel training environment named type 1 diabetes (T1D) simulation-to-reality (T1Dsim2real). Here, we conduct a stepwise validation experiment to validate the performance of the proposed control algorithm, which comprises in silico, retrospective-counterfactual studies, and preclinical studies using a T1D pig model. Results from the preclinical validation demonstrate the effectiveness of the proposed algorithm, with average time in target range of 70–180 mg/dL of 72.8 %, 73.8 %, 74.5 %, and 86.9 % across breakfast, lunch, dinner, and overnight fasting time, respectively. Thus, this study represents the first preclinical validation of a DRL-driven fully-AID algorithm in PwD, confirming the efficacy of the xgDRL model in preclinical settings.

{"title":"Generalized reinforcement learning control algorithm for fully automated insulin delivery system","authors":"Vega Pradana Rachim , Junyoung Yoo , Jaeyeon Lee , Yein Lee , Sung-Min Park","doi":"10.1016/j.eswa.2025.126909","DOIUrl":"10.1016/j.eswa.2025.126909","url":null,"abstract":"<div><div>A fully automated insulin delivery (Fully-AID) system is expected to provide the ultimate safety, comfort, and a sense of freedom for people living with diabetes (PwD). Previous studies have shown the potential of a deep reinforcement learning (DRL) model for fully-AID control algorithm in simulation environment. However, the practical implementation is still challenging due to the domain gaps between simulation and real world scenario. In this manuscript, we proposed a novel generalized control algorithm, called xgDRL, to realize a DRL-driven fully-AID system. The generalization of the proposed algorithm is achieved by our two main contributions that are introducing a novel concept of fully-AID context called total daily insulin (TDI) into the input of DRL model, and a novel training environment named type 1 diabetes (T1D) simulation-to-reality (T1Dsim2real). Here, we conduct a stepwise validation experiment to validate the performance of the proposed control algorithm, which comprises <em>in silico</em>, retrospective-counterfactual studies, and preclinical studies using a T1D pig model. Results from the preclinical validation demonstrate the effectiveness of the proposed algorithm, with average time in target range of 70–180 mg/dL of 72.8 %, 73.8 %, 74.5 %, and 86.9 % across breakfast, lunch, dinner, and overnight fasting time, respectively. Thus, this study represents the first preclinical validation of a DRL-driven fully-AID algorithm in PwD, confirming the efficacy of the xgDRL model in preclinical settings.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"274 ","pages":"Article 126909"},"PeriodicalIF":7.5,"publicationDate":"2025-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143474267","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}

引用次数: 0

Texture-embedded Generative Adversarial Nets for the synthesis of 3D pulmonary nodules computed tomography images

IF 7.5 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Expert Systems with Applications

Pub Date : 2025-02-19 DOI: 10.1016/j.eswa.2025.126860

Yi-Chang Chen , Ling-Ying Chiu , Chi-En Lee , Wei-Chieh Huang , Li-Wei Chen , Mong-Wei Lin , Ai-Su Yang , Ying-Zhen Ye , De-Xiang Ou , Yeun-Chung Chang , Chung-Ming Chen

Lung cancer is the second most common cancer and the leading cause of cancer-related death worldwide. Screening with low-dose computed tomography is crucial to detect early-stage lung cancer. Computer-aided diagnosis (CAD) can help clinicians to make diagnosis more quickly and more accurately. CAD based on deep learning algorithms is gaining attention. These algorithms rely on large amount of training data, which are barely available in the field of medical imaging, therefore data augmentation becomes essential. Generative Adversarial Nets (GAN) is an emerging solution for data augmentation and has been successfully used to generate realistic pulmonary nodules. In this study, we developed Texture-embedded GAN, which took the texture of nodule into consideration by introducing a loss function based on Gabor filters. We trained Texture-embedded GAN with images of 1075 nodule from the LIDC-IDRI dataset. Visual Turing Test showed that Texture-embedded GAN could generate images realistic enough to deceive expert radiologists. Data augmentation with Texture-embedded GAN improved the performance of ResNet-based classifier, which could distinguish benign and malignant nodules with 0.883 accuracy and 0.950 AUC. It was concluded that Texture-embedded GAN could generate realistic pulmonary nodules with sufficient diversity and was useful for data augmentation.

{"title":"Texture-embedded Generative Adversarial Nets for the synthesis of 3D pulmonary nodules computed tomography images","authors":"Yi-Chang Chen , Ling-Ying Chiu , Chi-En Lee , Wei-Chieh Huang , Li-Wei Chen , Mong-Wei Lin , Ai-Su Yang , Ying-Zhen Ye , De-Xiang Ou , Yeun-Chung Chang , Chung-Ming Chen","doi":"10.1016/j.eswa.2025.126860","DOIUrl":"10.1016/j.eswa.2025.126860","url":null,"abstract":"<div><div>Lung cancer is the second most common cancer and the leading cause of cancer-related death worldwide. Screening with low-dose computed tomography is crucial to detect early-stage lung cancer. Computer-aided diagnosis (CAD) can help clinicians to make diagnosis more quickly and more accurately. CAD based on deep learning algorithms is gaining attention. These algorithms rely on large amount of training data, which are barely available in the field of medical imaging, therefore data augmentation becomes essential. Generative Adversarial Nets (GAN) is an emerging solution for data augmentation and has been successfully used to generate realistic pulmonary nodules. In this study, we developed Texture-embedded GAN, which took the texture of nodule into consideration by introducing a loss function based on Gabor filters. We trained Texture-embedded GAN with images of 1075 nodule from the LIDC-IDRI dataset. Visual Turing Test showed that Texture-embedded GAN could generate images realistic enough to deceive expert radiologists. Data augmentation with Texture-embedded GAN improved the performance of ResNet-based classifier, which could distinguish benign and malignant nodules with 0.883 accuracy and 0.950 AUC. It was concluded that Texture-embedded GAN could generate realistic pulmonary nodules with sufficient diversity and was useful for data augmentation.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"274 ","pages":"Article 126860"},"PeriodicalIF":7.5,"publicationDate":"2025-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143474358","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}

引用次数: 0

Enhancing multi-step air quality prediction with deep learning using residual neural network and adaptive decomposition-based multi-objective optimization

IF 7.5 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Expert Systems with Applications

Pub Date : 2025-02-19 DOI: 10.1016/j.eswa.2025.126969

Kun Hu , Jinxing Che , Wenxin Xia , Yifan Xu , Yuerong Li

Accurate air quality prediction is crucial for health and ecology. However, existing studies often overlook the impact of data quality, feature extraction, external factors, and prediction uncertainty after data decomposition. To address this, we propose an enhanced multi-step air quality prediction approach using deep learning, incorporating residual neural networks and adaptive decomposition-based multi-objective optimization. This framework integrates meteorological factors and air pollutants, extracting trend and periodic features while ensuring smooth decomposition with minimal residuals. Training and prediction utilize a deep learning model based on residual networks, optimized with an improved arithmetic algorithm. Uncertainty prediction is implemented by modeling and sampling the prediction error. Experimental validation on data from Beijing, Shanghai, and Guangzhou demonstrates significant advantages over other models, confirming the reliability and accuracy of our framework in handling time series data and forecasting future trends. Additionally, uncertainty forecasting enhances forecast reliability and accuracy by describing the range of possible outcomes.

{"title":"Enhancing multi-step air quality prediction with deep learning using residual neural network and adaptive decomposition-based multi-objective optimization","authors":"Kun Hu , Jinxing Che , Wenxin Xia , Yifan Xu , Yuerong Li","doi":"10.1016/j.eswa.2025.126969","DOIUrl":"10.1016/j.eswa.2025.126969","url":null,"abstract":"<div><div>Accurate air quality prediction is crucial for health and ecology. However, existing studies often overlook the impact of data quality, feature extraction, external factors, and prediction uncertainty after data decomposition. To address this, we propose an enhanced multi-step air quality prediction approach using deep learning, incorporating residual neural networks and adaptive decomposition-based multi-objective optimization. This framework integrates meteorological factors and air pollutants, extracting trend and periodic features while ensuring smooth decomposition with minimal residuals. Training and prediction utilize a deep learning model based on residual networks, optimized with an improved arithmetic algorithm. Uncertainty prediction is implemented by modeling and sampling the prediction error. Experimental validation on data from Beijing, Shanghai, and Guangzhou demonstrates significant advantages over other models, confirming the reliability and accuracy of our framework in handling time series data and forecasting future trends. Additionally, uncertainty forecasting enhances forecast reliability and accuracy by describing the range of possible outcomes.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"273 ","pages":"Article 126969"},"PeriodicalIF":7.5,"publicationDate":"2025-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143453664","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}

引用次数: 0

ABUS-Net: Graph convolutional network with multi-scale features for breast cancer diagnosis using automated breast ultrasound

IF 7.5 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Expert Systems with Applications

Pub Date : 2025-02-19 DOI: 10.1016/j.eswa.2025.126978

Changyan Wang , Yuqing Guo , Haobo Chen , Qihui Guo , Haihao He , Lin Chen , Qi Zhang

Breast cancer is the leading cause of cancer-related deaths in women. Early screening helps with the treatment and recovery of breast cancer. The automated breast ultrasound (ABUS), as a standardized 3D breast ultrasound imaging technology, overcomes the limitations of traditional ultrasound, such as strong dependence on operator skills and poor reproducibility. At the same time, the coronal view provided by ABUS contains rich information that aids in the diagnosis of breast cancer. Therefore, how to effectively utilize the coronal features and spatial structural relationships is an urgent challenge. To address this issue, a graph convolutional network (GCN) with multi-scale features is proposed for breast cancer diagnosis using ABUS, named ABUS-Net. Unlike previous studies that relied on 3D patch techniques to represent tumor spatial features, our method focuses on the deep exploration of coronal features while using a GCN model to capture the spatial structural relationships of breast tumors. Specifically, we design a multi-scale feature extraction module to capture detailed information at different scales in the ABUS coronal sections, thereby enhancing the tumor feature representation. We then treat each slice containing tumor as a graph vertex, with the inherent spatial relationships between slices forming the edges. Finally, we employ the GCN model to classify the malignancy of the breast tumor. To validate the effectiveness and superiority of the model, we test it on both private and public datasets and compare it with other existing models. Experimental results highlight the potential utility of the proposed model in clinical practice.

{"title":"ABUS-Net: Graph convolutional network with multi-scale features for breast cancer diagnosis using automated breast ultrasound","authors":"Changyan Wang , Yuqing Guo , Haobo Chen , Qihui Guo , Haihao He , Lin Chen , Qi Zhang","doi":"10.1016/j.eswa.2025.126978","DOIUrl":"10.1016/j.eswa.2025.126978","url":null,"abstract":"<div><div>Breast cancer is the leading cause of cancer-related deaths in women. Early screening helps with the treatment and recovery of breast cancer. The automated breast ultrasound (ABUS), as a standardized 3D breast ultrasound imaging technology, overcomes the limitations of traditional ultrasound, such as strong dependence on operator skills and poor reproducibility. At the same time, the coronal view provided by ABUS contains rich information that aids in the diagnosis of breast cancer. Therefore, how to effectively utilize the coronal features and spatial structural relationships is an urgent challenge. To address this issue, a graph convolutional network (GCN) with multi-scale features is proposed for breast cancer diagnosis using ABUS, named ABUS-Net. Unlike previous studies that relied on 3D patch techniques to represent tumor spatial features, our method focuses on the deep exploration of coronal features while using a GCN model to capture the spatial structural relationships of breast tumors. Specifically, we design a multi-scale feature extraction module to capture detailed information at different scales in the ABUS coronal sections, thereby enhancing the tumor feature representation. We then treat each slice containing tumor as a graph vertex, with the inherent spatial relationships between slices forming the edges. Finally, we employ the GCN model to classify the malignancy of the breast tumor. To validate the effectiveness and superiority of the model, we test it on both private and public datasets and compare it with other existing models. Experimental results highlight the potential utility of the proposed model in clinical practice.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"273 ","pages":"Article 126978"},"PeriodicalIF":7.5,"publicationDate":"2025-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143464074","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}

引用次数: 0

Modified γ-operator technique for intuitionistic fuzzy multi-objective nonlinear programming problems with application in agriculture

IF 7.5 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Expert Systems with Applications

Pub Date : 2025-02-19 DOI: 10.1016/j.eswa.2025.126874

Shubhpreet Kaur, Sumati Mahajan, Abhishek Chauhan

In recent years, intuitionistic fuzzy theory has gained significant attention for its ability to handle uncertainty through both membership and non-membership degrees. This paper presents a novel modification to

γ

-operator technique by introducing distinct

γ

-operators for membership and non-membership functions to tackle intuitionistic fuzzy environment. The proposed technique is rigorously validated through the proof of relevant theorems that demonstrate its capability to derive efficient solutions for multi-objective nonlinear programming problems, where all parameters are represented as triangular intuitionistic fuzzy numbers. To elucidate the proposed technique, an illustrative example is presented. Furthermore, a comparative study with existing techniques is conducted, which highlights the superior performance of the proposed method. Finally, an application in the agriculture sector demonstrates the practical relevance and effectiveness of the proposed method in real-world scenarios.

{"title":"Modified γ-operator technique for intuitionistic fuzzy multi-objective nonlinear programming problems with application in agriculture","authors":"Shubhpreet Kaur, Sumati Mahajan, Abhishek Chauhan","doi":"10.1016/j.eswa.2025.126874","DOIUrl":"10.1016/j.eswa.2025.126874","url":null,"abstract":"<div><div>In recent years, intuitionistic fuzzy theory has gained significant attention for its ability to handle uncertainty through both membership and non-membership degrees. This paper presents a novel modification to <span><math><mi>γ</mi></math></span>-operator technique by introducing distinct <span><math><mi>γ</mi></math></span>-operators for membership and non-membership functions to tackle intuitionistic fuzzy environment. The proposed technique is rigorously validated through the proof of relevant theorems that demonstrate its capability to derive efficient solutions for multi-objective nonlinear programming problems, where all parameters are represented as triangular intuitionistic fuzzy numbers. To elucidate the proposed technique, an illustrative example is presented. Furthermore, a comparative study with existing techniques is conducted, which highlights the superior performance of the proposed method. Finally, an application in the agriculture sector demonstrates the practical relevance and effectiveness of the proposed method in real-world scenarios.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"273 ","pages":"Article 126874"},"PeriodicalIF":7.5,"publicationDate":"2025-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143471193","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}

引用次数: 0

From grids to pseudo-regions: Dynamic memory augmented image captioning with dual relation transformer

IF 7.5 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Expert Systems with Applications

Pub Date : 2025-02-19 DOI: 10.1016/j.eswa.2025.126850

Wei Zhou, Weitao Jiang, Zhijie Zheng, Jianchao Li, Tao Su, Haifeng Hu

Image captioning aims to automatically generate a description in natural language for a given image. The existing methods typically exploit grid-level or region-level features to encode visual information. However, extracting region features by an object detector is computationally expensive and inflexible, while region features are criticized for lacking fine-grained details and background information. Besides, current Transformer-based captioning models only focus on the pairwise similarity of token features, which makes it difficult to fully understand the complex scene relationships in images. To address these issues, we introduce a novel Dual Relation Transformer (DRTran) model that can be trained end-to-end. Concretely, in the encoding phase, we first adopt a clustering algorithm to generate pseudo-region features, which does not need to make additional expensive annotations to train object detector. Then, in order to combine the advantages of grid and pseudo-region features, we design a new dual relation enhancement (DRE) encoder to capture the correlation between objects from two different visual features. Furthermore, we devise a novel dynamic memory (DM) module to learn prior knowledge with external dynamic memory vectors. By adding prior knowledge in visual relationship modeling, the model learns complex scene representations to improve caption accuracy. During the decoding stage, we design a new cross-modal attention fusion (CAF) module in the language decoder to adaptively decide the attention weights of enhanced grid and pseudo-region features at each time step. Extensive experiments on the MS-COCO and Flickr30K datasets demonstrate that our DRTran model performs better than current image captioning methods.

{"title":"From grids to pseudo-regions: Dynamic memory augmented image captioning with dual relation transformer","authors":"Wei Zhou, Weitao Jiang, Zhijie Zheng, Jianchao Li, Tao Su, Haifeng Hu","doi":"10.1016/j.eswa.2025.126850","DOIUrl":"10.1016/j.eswa.2025.126850","url":null,"abstract":"<div><div>Image captioning aims to automatically generate a description in natural language for a given image. The existing methods typically exploit grid-level or region-level features to encode visual information. However, extracting region features by an object detector is computationally expensive and inflexible, while region features are criticized for lacking fine-grained details and background information. Besides, current Transformer-based captioning models only focus on the pairwise similarity of token features, which makes it difficult to fully understand the complex scene relationships in images. To address these issues, we introduce a novel Dual Relation Transformer (DRTran) model that can be trained end-to-end. Concretely, in the encoding phase, we first adopt a clustering algorithm to generate pseudo-region features, which does not need to make additional expensive annotations to train object detector. Then, in order to combine the advantages of grid and pseudo-region features, we design a new dual relation enhancement (DRE) encoder to capture the correlation between objects from two different visual features. Furthermore, we devise a novel dynamic memory (DM) module to learn prior knowledge with external dynamic memory vectors. By adding prior knowledge in visual relationship modeling, the model learns complex scene representations to improve caption accuracy. During the decoding stage, we design a new cross-modal attention fusion (CAF) module in the language decoder to adaptively decide the attention weights of enhanced grid and pseudo-region features at each time step. Extensive experiments on the MS-COCO and Flickr30K datasets demonstrate that our DRTran model performs better than current image captioning methods.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"273 ","pages":"Article 126850"},"PeriodicalIF":7.5,"publicationDate":"2025-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143464363","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}

引用次数: 0

Study on field strength prediction using different models on time series from urban continuous RF-EMF monitoring

IF 7.5 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Expert Systems with Applications

Pub Date : 2025-02-19 DOI: 10.1016/j.eswa.2025.126963

Xinwei Song , Wenjun Feng , Chen Yang , Nikola Djuric , Dragan Kljajic , Snezana Djuric

State-of-the-art electromagnetic field (EMF) monitoring networks, such as the latest one – the Serbian EMF RATEL system, are able to provide continuous and daily monitoring of radio-frequency (RF) EMF levels, which is especially important for urban areas where people may spend many hours and, consequently, experience increased sensitivity to RF-EMF exposure. By generating considerable time series sets of RF-EMF data, these monitoring networks are initiating a new research topic – near-future RF-EMF prediction, which is valuable for a number of public health activities, from supplementing EMF monitoring in high-risk areas, to proactively reducing exposure times, and towards advancing pre-testing of EMF compliance. This paper investigates the impact of different models on the prediction of field strength in urban environments, where Seasonal Auto-Regressive Integrated Moving Average (SARIMA), Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM), Extreme Learning Machine (ELM), Partial Least Squares Regression (PLS) and Transformer models are considered. The prediction performance of each model is analyzed on a case study of EMF-sensitive areas in the Serbian city of Novi Sad, i.e., two kindergartens and an elementary school; however, the established framework has strong potential for generalization to other urban environments. Based on two-year long monitoring data sets, a comprehensive comparison of the six models on prediction accuracy, performance degradation rate, extreme value prediction accuracy, and training time is made, showing that the PLS model outperforms other models in predicting EMF exposure. This preliminary study may be a valuable reference for large-scale deployment in real-time monitoring systems for public health protection and may trigger additional research on this ultimate EMF topic.

{"title":"Study on field strength prediction using different models on time series from urban continuous RF-EMF monitoring","authors":"Xinwei Song , Wenjun Feng , Chen Yang , Nikola Djuric , Dragan Kljajic , Snezana Djuric","doi":"10.1016/j.eswa.2025.126963","DOIUrl":"10.1016/j.eswa.2025.126963","url":null,"abstract":"<div><div>State-of-the-art electromagnetic field (EMF) monitoring networks, such as the latest one – the Serbian EMF RATEL system, are able to provide continuous and daily monitoring of radio-frequency (RF) EMF levels, which is especially important for urban areas where people may spend many hours and, consequently, experience increased sensitivity to RF-EMF exposure. By generating considerable time series sets of RF-EMF data, these monitoring networks are initiating a new research topic – near-future RF-EMF prediction, which is valuable for a number of public health activities, from supplementing EMF monitoring in high-risk areas, to proactively reducing exposure times, and towards advancing pre-testing of EMF compliance. This paper investigates the impact of different models on the prediction of field strength in urban environments, where Seasonal Auto-Regressive Integrated Moving Average (SARIMA), Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM), Extreme Learning Machine (ELM), Partial Least Squares Regression (PLS) and Transformer models are considered. The prediction performance of each model is analyzed on a case study of EMF-sensitive areas in the Serbian city of Novi Sad, i.e., two kindergartens and an elementary school; however, the established framework has strong potential for generalization to other urban environments. Based on two-year long monitoring data sets, a comprehensive comparison of the six models on prediction accuracy, performance degradation rate, extreme value prediction accuracy, and training time is made, showing that the PLS model outperforms other models in predicting EMF exposure. This preliminary study may be a valuable reference for large-scale deployment in real-time monitoring systems for public health protection and may trigger additional research on this ultimate EMF topic.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"274 ","pages":"Article 126963"},"PeriodicalIF":7.5,"publicationDate":"2025-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143488619","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}

引用次数: 0

Invariant semantic domain generalization shuffle network for cross-scene hyperspectral image classification

IF 7.5 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Expert Systems with Applications

Pub Date : 2025-02-19 DOI: 10.1016/j.eswa.2025.126818

Jingpeng Gao, Xiangyu Ji, Fang Ye, Geng Chen

Cross-scene hyperspectral image classification is currently receiving widespread attention. However, domain adaptation-based methods usually perform domain alignment by accessing specific target scenes during training and require retraining for new scenes. In contrast, domain generalization only trains using the source domain and then gradually generalizes to unseen domains. However, existing methods based on domain generalization ignore the impact of domain invariant semantics on the invariant representation of the domain. To solve the above problem, an invariant semantic domain generalization shuffle network for cross-scene hyperspectral image classification is proposed, which follows a framework on the generative adversarial network. Feature style covariance in style and content randomization generator with invariant semantic features is designed to safely extend the style and content of features without changing the domain invariant semantics. We proposed a spatial shuffling discriminator, which can reduce the impact of special spatial relationships within the domain on class semantics. In addition, we proposed a dual sampling direct adversarial contrastive learning strategy. It uses a dual sampling in two-stage training design to prevent the model from lazily entering the local nash equilibrium point. And based on dual sampling, directly adversarial contrastive learning using clearer contrastive samples is used to reduce the difficulty of network training. We conduct extensive experiments on four datasets and demonstrate that the proposed method outperforms other current domain generalization methods. The code will be open source at https://github.com/jixiangyu0501/ISDGS.

{"title":"Invariant semantic domain generalization shuffle network for cross-scene hyperspectral image classification","authors":"Jingpeng Gao, Xiangyu Ji, Fang Ye, Geng Chen","doi":"10.1016/j.eswa.2025.126818","DOIUrl":"10.1016/j.eswa.2025.126818","url":null,"abstract":"<div><div>Cross-scene hyperspectral image classification is currently receiving widespread attention. However, domain adaptation-based methods usually perform domain alignment by accessing specific target scenes during training and require retraining for new scenes. In contrast, domain generalization only trains using the source domain and then gradually generalizes to unseen domains. However, existing methods based on domain generalization ignore the impact of domain invariant semantics on the invariant representation of the domain. To solve the above problem, an invariant semantic domain generalization shuffle network for cross-scene hyperspectral image classification is proposed, which follows a framework on the generative adversarial network. Feature style covariance in style and content randomization generator with invariant semantic features is designed to safely extend the style and content of features without changing the domain invariant semantics. We proposed a spatial shuffling discriminator, which can reduce the impact of special spatial relationships within the domain on class semantics. In addition, we proposed a dual sampling direct adversarial contrastive learning strategy. It uses a dual sampling in two-stage training design to prevent the model from lazily entering the local nash equilibrium point. And based on dual sampling, directly adversarial contrastive learning using clearer contrastive samples is used to reduce the difficulty of network training. We conduct extensive experiments on four datasets and demonstrate that the proposed method outperforms other current domain generalization methods. The code will be open source at <span><span>https://github.com/jixiangyu0501/ISDGS</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"273 ","pages":"Article 126818"},"PeriodicalIF":7.5,"publicationDate":"2025-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143453671","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}

引用次数: 0