Pub Date : 2025-02-01DOI: 10.1016/j.asoc.2024.112654
Prativa Agarwalla , Sumitra Mukhopadhyay
High dimensional microarray cancer datasets contain thousands of genes with a very few numbers of samples. High class imbalance, presence of noisy and redundant genes and overlapping nature of extracted features among different disease classes deteriorate the disease prediction accuracy. An intelligent collaborative filtering (ICF) assisted and hamming distance guided multi-objective swarm intelligence framework (HIMS) is proposed for efficient selection of optimal gene set for disease identification. In the framework, first intelligent collaborative filtering (ICF) has been introduced to improve the prediction ability which combines the features from different feature selection tools. Then, a multi-objective multi-population search (MOMPS) algorithm has been proposed which contributes as a core part of HIMS. It generates more diversified solutions by avoiding local trapping. Hamming distance operator has been applied here as an alternative of sorting mechanism for the selection of Pareto optimal solutions. It also helps to reduce the computational complexity. Along with that, a time-varying U-shaped function is introduced for the binary conversion process for feature selection. Extensive experiments were conducted on 16 different single and multi-class datasets to study the efficacy of HIMS. The experimental results show that HIMS performs favorably well in comparison with other existing techniques with fewer numbers of genes.
{"title":"Gene expression selection for cancer classification using intelligent collaborative filtering and hamming distance guided multi-objective swarm optimization","authors":"Prativa Agarwalla , Sumitra Mukhopadhyay","doi":"10.1016/j.asoc.2024.112654","DOIUrl":"10.1016/j.asoc.2024.112654","url":null,"abstract":"<div><div>High dimensional microarray cancer datasets contain thousands of genes with a very few numbers of samples. High class imbalance, presence of noisy and redundant genes and overlapping nature of extracted features among different disease classes deteriorate the disease prediction accuracy. An intelligent collaborative filtering (ICF) assisted and hamming distance guided multi-objective swarm intelligence framework (HIMS) is proposed for efficient selection of optimal gene set for disease identification. In the framework, first intelligent collaborative filtering (ICF) has been introduced to improve the prediction ability which combines the features from different feature selection tools. Then, a multi-objective multi-population search (MOMPS) algorithm has been proposed which contributes as a core part of HIMS. It generates more diversified solutions by avoiding local trapping. Hamming distance operator has been applied here as an alternative of sorting mechanism for the selection of Pareto optimal solutions. It also helps to reduce the computational complexity. Along with that, a time-varying U-shaped function is introduced for the binary conversion process for feature selection. Extensive experiments were conducted on 16 different single and multi-class datasets to study the efficacy of HIMS. The experimental results show that HIMS performs favorably well in comparison with other existing techniques with fewer numbers of genes.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"170 ","pages":"Article 112654"},"PeriodicalIF":7.2,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143213092","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}
Pub Date : 2025-02-01DOI: 10.1016/j.asoc.2025.112697
Libao Deng , Yixuan Qiu , Yuanzhu Di , Lili Zhang
Flexible job shop scheduling (FJSP) stands as one of the most pivotal scheduling problems, attracting considerable research efforts aimed at discovering improved solutions. The distributed variant of FJSP, which extends the problem’s scope, has garnered substantial interest among scholars. Recognizing the inherent limitation in machine endurance and the inevitable degradation with accumulating workloads, the significance of preventive maintenance in enhancing machine reliability is emphasized and ensuring process control. This study endeavors to concurrently optimize three key metrics: makespan, total energy consumption, and maintenance cost. To this end, a knowledge-driven memetic algorithm is tailored specifically to the problem’s characteristics. Our approach commences with a hybrid initialization incorporating eight strategic approaches, crafted to address the factory assignment, operation sequence, and machine selection subproblems, thereby yielding an initial population characterized by high quality and diversity. Subsequently, genetic operators are employed to generate offspring, wherein elite segments from exemplary solutions are selectively inherited during crossover. A two-stage mutation mechanism is introduced to foster the emergence of novel individuals. Finally, three tailored local search strategies are executed, striking a balance between exploration and exploitation.Comprehensive experimental findings emphasize the superior performance of the proposed algorithm in addressing the pertinent problem. The experimental results presented in this paper indicate increases of 20% and 141% in Hypervolume (HV) values and Metric for Diversity (DM) values, respectively, while the reduction in Inverted Generation Distance (IGD) values amounts to 85%, thereby demonstrating the effectiveness of our proposed methodology.
{"title":"A knowledge-driven memetic algorithm for distributed green flexible job shop scheduling considering the endurance of machines","authors":"Libao Deng , Yixuan Qiu , Yuanzhu Di , Lili Zhang","doi":"10.1016/j.asoc.2025.112697","DOIUrl":"10.1016/j.asoc.2025.112697","url":null,"abstract":"<div><div>Flexible job shop scheduling (FJSP) stands as one of the most pivotal scheduling problems, attracting considerable research efforts aimed at discovering improved solutions. The distributed variant of FJSP, which extends the problem’s scope, has garnered substantial interest among scholars. Recognizing the inherent limitation in machine endurance and the inevitable degradation with accumulating workloads, the significance of preventive maintenance in enhancing machine reliability is emphasized and ensuring process control. This study endeavors to concurrently optimize three key metrics: makespan, total energy consumption, and maintenance cost. To this end, a knowledge-driven memetic algorithm is tailored specifically to the problem’s characteristics. Our approach commences with a hybrid initialization incorporating eight strategic approaches, crafted to address the factory assignment, operation sequence, and machine selection subproblems, thereby yielding an initial population characterized by high quality and diversity. Subsequently, genetic operators are employed to generate offspring, wherein elite segments from exemplary solutions are selectively inherited during crossover. A two-stage mutation mechanism is introduced to foster the emergence of novel individuals. Finally, three tailored local search strategies are executed, striking a balance between exploration and exploitation.Comprehensive experimental findings emphasize the superior performance of the proposed algorithm in addressing the pertinent problem. The experimental results presented in this paper indicate increases of 20% and 141% in Hypervolume (HV) values and Metric for Diversity (DM) values, respectively, while the reduction in Inverted Generation Distance (IGD) values amounts to 85%, thereby demonstrating the effectiveness of our proposed methodology.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"170 ","pages":"Article 112697"},"PeriodicalIF":7.2,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143213112","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}
Pub Date : 2025-02-01DOI: 10.1016/j.asoc.2024.112642
Qian Hu , Jun Zhang , Jusheng Mi , Zhong Yuan , Meizheng Li
Outlier detection is an attractive research area in data mining, which is intended to find out the few data objects that are abnormal to the normal data set. Formal concept analysis is an efficacious mathematical tool to perform data analysis and processing. Three-way concepts contain both information of co-having and co-not-having, and reflect the correlation among objects (attributes). Information entropy reflects the degree of uncertainty of the system. Information entropy-based outlier detection methods have been widely studied and have shown excellent performance, but most current information entropy-based methods contain parameters, which leads to detection results are sensitive to parameters settings and taking longer detection times. Aiming at this deficiency, this paper constructs a three-way concept-based information entropy outlier detection method. Firstly, the information entropy of the formal context is defined by utilizing three-way granular concepts, and then the relative entropy of each object is defined. According to it, the relative cardinality-based outlier degree of each object is given, and then the outlier factor of the object is defined by combining with the relative entropy. Then the three-way concept information entropy-based outlier factor is presented and the associated algorithm is proposed. Finally, the effectiveness and efficiency of the proposed algorithm is verified on a public dataset.
{"title":"TIEOD: Three-way concept-based information entropy for outlier detection","authors":"Qian Hu , Jun Zhang , Jusheng Mi , Zhong Yuan , Meizheng Li","doi":"10.1016/j.asoc.2024.112642","DOIUrl":"10.1016/j.asoc.2024.112642","url":null,"abstract":"<div><div>Outlier detection is an attractive research area in data mining, which is intended to find out the few data objects that are abnormal to the normal data set. Formal concept analysis is an efficacious mathematical tool to perform data analysis and processing. Three-way concepts contain both information of co-having and co-not-having, and reflect the correlation among objects (attributes). Information entropy reflects the degree of uncertainty of the system. Information entropy-based outlier detection methods have been widely studied and have shown excellent performance, but most current information entropy-based methods contain parameters, which leads to detection results are sensitive to parameters settings and taking longer detection times. Aiming at this deficiency, this paper constructs a three-way concept-based information entropy outlier detection method. Firstly, the information entropy of the formal context is defined by utilizing three-way granular concepts, and then the relative entropy of each object is defined. According to it, the relative cardinality-based outlier degree of each object is given, and then the outlier factor of the object is defined by combining with the relative entropy. Then the three-way concept information entropy-based outlier factor is presented and the associated algorithm is proposed. Finally, the effectiveness and efficiency of the proposed algorithm is verified on a public dataset.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"170 ","pages":"Article 112642"},"PeriodicalIF":7.2,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143213398","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}
Pub Date : 2025-02-01DOI: 10.1016/j.asoc.2024.112559
Harish Sharma , Edgar Galván , Peter Mooney
A Parallel Genetic Algorithm (PGA) specifically designed for multi-criteria vehicle routing is described in this work. The algorithm aims to enhance the existing routing methods by offering users the ability to choose their preferred path from a set of optimal paths optimised on multiple objectives. The objectives are optimised using a novel fitness metric that prioritises minimising path length while also maximising access to specific amenities such as pubs, hotels, and charging stations. The developed approach, called Parallel Optimal-route Search (POS), follows a hybrid model using both global parallelisation and island-based approaches. A Loop-Free Path-Composer (LFPC) is described and this genetic operator generates new paths for evaluation and is shown to yield a more diverse set of solutions in contrast to other commonly used approaches, such as Node Based Crossover and Path Mutation (NBCPM). Our approach is validated on highly complex, large-scale real-world road networks, with sizes ranging from 3,000 and 10,000 nodes. We present a systematic study comparing the performance of our proposed LFPC operator against the traditional NBCPM operators. Additionally, we evaluate the effectiveness of our proposed POS algorithm in comparison to the well-known Non-dominated Sorting Genetic Algorithm II and III.
{"title":"A parallel genetic algorithm for multi-criteria path routing on complex real-world road networks","authors":"Harish Sharma , Edgar Galván , Peter Mooney","doi":"10.1016/j.asoc.2024.112559","DOIUrl":"10.1016/j.asoc.2024.112559","url":null,"abstract":"<div><div>A Parallel Genetic Algorithm (PGA) specifically designed for multi-criteria vehicle routing is described in this work. The algorithm aims to enhance the existing routing methods by offering users the ability to choose their preferred path from a set of optimal paths optimised on multiple objectives. The objectives are optimised using a novel fitness metric that prioritises minimising path length while also maximising access to specific amenities such as pubs, hotels, and charging stations. The developed approach, called Parallel Optimal-route Search (POS), follows a hybrid model using both global parallelisation and island-based approaches. A Loop-Free Path-Composer (LFPC) is described and this genetic operator generates new paths for evaluation and is shown to yield a more diverse set of solutions in contrast to other commonly used approaches, such as Node Based Crossover and Path Mutation (NBCPM). Our approach is validated on highly complex, large-scale real-world road networks, with sizes ranging from 3,000 and 10,000 nodes. We present a systematic study comparing the performance of our proposed LFPC operator against the traditional NBCPM operators. Additionally, we evaluate the effectiveness of our proposed POS algorithm in comparison to the well-known Non-dominated Sorting Genetic Algorithm II and III.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"170 ","pages":"Article 112559"},"PeriodicalIF":7.2,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143213074","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}
Pub Date : 2025-02-01DOI: 10.1016/j.asoc.2024.112673
Sajid Hussain , Xu Songhua , Muhammad Usman Aslam , Muhammad Waqas , Fida Hussain
The Monkeypox virus (Mpoxv), characterized by its distinct vesiculopustular rash, has re-emerged as a significant zoonotic pathogen, posing severe public health risks and potential bioterrorism threats. Although less virulent than smallpox, the persistence of Mpoxv infections necessitates advanced diagnostic tools and proactive mitigation strategies. Dermatological virological imaging is significant for automatic Mpoxv detection and classification, yet its fidelity is often compromised by low-resolution data, mainly in the incipient stages of the infection. Conventional deep learning models mostly struggle to capture higher-order dependencies and complex feature interactions within virological images, leading to suboptimal outcomes. An advanced hybrid hypergraph convolutional neural networks (HGCNs) architecture is introduced in response. In this architecture hypergraph effectively models intricate correlations and enables the detect subtle patterns. At the same time, CNN components contribute robust feature extraction, refined through relational modeling, leads optimal detection and classification of Mpoxv infection. The HGCNs were trained and validated using two different validation approaches, including the Holdout method (HM) and a stratified 3-fold cross-validation (3-FCV), yielding HM accuracy of 0.9888, precision of 0.9813, recall of 0.9958, F1 Score of 0.9885, specificity of 0.9890, Micro AUC of 0.9892, and an average time per epoch of 0.5512 s, while 3-FCV achieved an average accuracy of 0.9917, precision of 0.9931, recall of 0.9912, F1 score of 0.9922, specificity of 0.9941, Micro AUC of 0.9903, and an average time per epoch of 0.6151 s. Furthermore, the use of Grad-CAM facilitates precise localization of infected regions within the images. The performance highlights the proposed model’s effectiveness as a powerful tool in computational virology, delivering high accuracy and interpretable diagnostics for Mpoxv infections.
Data availability
The dataset is freely available online.
{"title":"Hypergraph convolutional neural networks for clinical diagnosis of monkeypox infections using skin virological images","authors":"Sajid Hussain , Xu Songhua , Muhammad Usman Aslam , Muhammad Waqas , Fida Hussain","doi":"10.1016/j.asoc.2024.112673","DOIUrl":"10.1016/j.asoc.2024.112673","url":null,"abstract":"<div><div>The Monkeypox virus (Mpoxv), characterized by its distinct vesiculopustular rash, has re-emerged as a significant zoonotic pathogen, posing severe public health risks and potential bioterrorism threats. Although less virulent than smallpox, the persistence of Mpoxv infections necessitates advanced diagnostic tools and proactive mitigation strategies. Dermatological virological imaging is significant for automatic Mpoxv detection and classification, yet its fidelity is often compromised by low-resolution data, mainly in the incipient stages of the infection. Conventional deep learning models mostly struggle to capture higher-order dependencies and complex feature interactions within virological images, leading to suboptimal outcomes. An advanced hybrid hypergraph convolutional neural networks (HGCNs) architecture is introduced in response. In this architecture hypergraph effectively models intricate correlations and enables the detect subtle patterns. At the same time, CNN components contribute robust feature extraction, refined through relational modeling, leads optimal detection and classification of Mpoxv infection. The HGCNs were trained and validated using two different validation approaches, including the Holdout method (HM) and a stratified 3-fold cross-validation (3-FCV), yielding HM accuracy of 0.9888, precision of 0.9813, recall of 0.9958, F1 Score of 0.9885, specificity of 0.9890, Micro AUC of 0.9892, and an average time per epoch of 0.5512 s, while 3-FCV achieved an average accuracy of 0.9917, precision of 0.9931, recall of 0.9912, F1 score of 0.9922, specificity of 0.9941, Micro AUC of 0.9903, and an average time per epoch of 0.6151 s. Furthermore, the use of Grad-CAM facilitates precise localization of infected regions within the images. The performance highlights the proposed model’s effectiveness as a powerful tool in computational virology, delivering high accuracy and interpretable diagnostics for Mpoxv infections.</div></div><div><h3>Data availability</h3><div>The dataset is freely available online.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"170 ","pages":"Article 112673"},"PeriodicalIF":7.2,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143213076","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}
Pub Date : 2025-02-01DOI: 10.1016/j.asoc.2025.112717
Hengdong Zhu , Baoshuo Kan , Yong Li , Enliang Yan , Heng Weng , Fu Lee Wang , Tianyong Hao
Fuzzy clustering is a simple but efficient clustering method, which aims to deal with ambiguous and overlapping data classification boundaries and provide detailed membership degree information. However, the complex structure of high-dimensional data in the real world easily causes the inefficiency of widely used distance metrics. In addition, traditional fuzzy clustering methods cannot utilize supervision information to guide membership degree matrix learning, which limits clustering performance. In this paper, we propose a new semi-supervised fuzzy clustering method based on latent representation learning and information fusion. Specifically, our method utilizes a deep autoencoder to learn the underlying low-dimensional structure of data, thereby mitigating the impact of redundant features. Moreover, a semi-supervised constraint term based on information fusion is designed to make full use of prior knowledge to supervise the clustering process. On the one hand, a new constraint distance is proposed by leveraging pairwise constraint information to re-evaluate the membership degree of data. On the other hand, the semi-supervised constraint term is constructed based on label information to guide the membership degree matrix learning. Comprehensive experiments on a variety of standard datasets show that our method achieves better performance compared with state-of-the-art baseline methods, demonstrating the effectiveness of the proposed method in fuzzy clustering.
{"title":"A new semi-supervised fuzzy clustering method based on latent representation learning and information fusion","authors":"Hengdong Zhu , Baoshuo Kan , Yong Li , Enliang Yan , Heng Weng , Fu Lee Wang , Tianyong Hao","doi":"10.1016/j.asoc.2025.112717","DOIUrl":"10.1016/j.asoc.2025.112717","url":null,"abstract":"<div><div>Fuzzy clustering is a simple but efficient clustering method, which aims to deal with ambiguous and overlapping data classification boundaries and provide detailed membership degree information. However, the complex structure of high-dimensional data in the real world easily causes the inefficiency of widely used distance metrics. In addition, traditional fuzzy clustering methods cannot utilize supervision information to guide membership degree matrix learning, which limits clustering performance. In this paper, we propose a new semi-supervised fuzzy clustering method based on latent representation learning and information fusion. Specifically, our method utilizes a deep autoencoder to learn the underlying low-dimensional structure of data, thereby mitigating the impact of redundant features. Moreover, a semi-supervised constraint term based on information fusion is designed to make full use of prior knowledge to supervise the clustering process. On the one hand, a new constraint distance is proposed by leveraging pairwise constraint information to re-evaluate the membership degree of data. On the other hand, the semi-supervised constraint term is constructed based on label information to guide the membership degree matrix learning. Comprehensive experiments on a variety of standard datasets show that our method achieves better performance compared with state-of-the-art baseline methods, demonstrating the effectiveness of the proposed method in fuzzy clustering.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"170 ","pages":"Article 112717"},"PeriodicalIF":7.2,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143213214","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}
Pub Date : 2025-02-01DOI: 10.1016/j.asoc.2025.112703
Gitanjali Ganpatrao Nikam, Jayanta Kumar Ghosh
The information of interest to any geospatial application requires being extracted from Topographic maps (TMs). Extracting information from topographic map represents one of the major bottlenecks due to complex distribution of geographic elements and highly interconnected nature of map features. The need for automated topographic map understanding arose because current methods for the information extraction are not adequate and an automated method is necessary in terms of time and economic efficiency.
This work reports on an implementation of Topographic map understanding system to extract spatial information and to provide this information in machine-readable data formats to preserve the digital repository. This data can be used for analytical purposes required in generation of. The paper presents Indian topographic map understanding system (ITMUS) that is characterized by the human mentation and learning capabilities. The ITMUS is comprised of image processing routines, Structure feature descriptors, and adaptive Neuro-fuzzy inference system. The Fuzzy inferencing has been implemented using Sugeno model which utilizes the initial crude domain knowledge about the map legends. Further, system has been trained for various sample regions selected from Open Series Map (OSM) Indian topographic maps. Results of implementation are evaluated against reference data of Survey of India and manual recognition. It has been found that the overall recognition rate of the system is 90.91 %. Further, the system’s overall accuracy is determined to be 92.77 %.
{"title":"Neuro-fuzzy based Indian Topographic map understanding system","authors":"Gitanjali Ganpatrao Nikam, Jayanta Kumar Ghosh","doi":"10.1016/j.asoc.2025.112703","DOIUrl":"10.1016/j.asoc.2025.112703","url":null,"abstract":"<div><div>The information of interest to any geospatial application requires being extracted from Topographic maps (TMs). Extracting information from topographic map represents one of the major bottlenecks due to complex distribution of geographic elements and highly interconnected nature of map features. The need for automated topographic map understanding arose because current methods for the information extraction are not adequate and an automated method is necessary in terms of time and economic efficiency.</div><div>This work reports on an implementation of Topographic map understanding system to extract spatial information and to provide this information in machine-readable data formats to preserve the digital repository. This data can be used for analytical purposes required in generation of. The paper presents Indian topographic map understanding system (ITMUS) that is characterized by the human mentation and learning capabilities. The ITMUS is comprised of image processing routines, Structure feature descriptors, and adaptive Neuro-fuzzy inference system. The Fuzzy inferencing has been implemented using Sugeno model which utilizes the initial crude domain knowledge about the map legends. Further, system has been trained for various sample regions selected from Open Series Map (OSM) Indian topographic maps. Results of implementation are evaluated against reference data of Survey of India and manual recognition. It has been found that the overall recognition rate of the system is 90.91 %. Further, the system’s overall accuracy is determined to be 92.77 %.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"170 ","pages":"Article 112703"},"PeriodicalIF":7.2,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143213215","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}
Pub Date : 2025-02-01DOI: 10.1016/j.asoc.2024.112416
Ali Nikhalat-Jahromi, Ali Mohammad Saghiri, Mohammad Reza Meybodi
Learning Automaton (LA), a branch of reinforcement learning, initially began with the Fixed Structure Learning Automaton (FSLA) family and was later expanded to include the Variable Structure Learning Automaton (VSLA) family. Tuning the depth of the FSLA in complex environments has long been a challenging task, significantly limiting the ability to effectively navigate the exploration–exploitation dilemma. No solution has been found for this open problem yet. This study addresses this issue by introducing a novel hybrid learning automaton model called Asymmetric Variable Depth Hybrid Learning Automaton (AVDHLA). The AVDHLA model intelligently learns the depth of fixed structure LA in an autonomous manner by combining the from the FSLA class and Variable Action Set LA (VASLA) from the VSLA class. Computer simulations are conducted to validate the proposed model in diverse environments, including both stationary and non-stationary (Markovian switching and State-dependent) scenarios. Performance evaluation is based on predefined metrics, such as total number of rewards (TNR) and action switching (TNAS). Statistical tests indicate that across both stationary and non-stationary environments, the AVDHLA consistently outperforms the in terms of TNR and TNAS across the majority of experiments. Moreover, the AVDHLA model is applied in two key applications. Firstly, it is used to defend against the selfish mining attack in Bitcoin and is compared with the well-known tie-breaking mechanism. Simulation results consistently demonstrate that our proposed method increases the threshold for successful selfish mining attacks from 25% to 40%. Secondly, the AVDHLA model has been applied to develop a novel learning automaton-based recommendation system. The results demonstrate the superiority of the proposed method in terms of the Click-Through Rate (CTR) and Precision compared to previous approaches.
{"title":"Asymmetric variable depth learning automaton and its application in defending against selfish mining attacks on bitcoin","authors":"Ali Nikhalat-Jahromi, Ali Mohammad Saghiri, Mohammad Reza Meybodi","doi":"10.1016/j.asoc.2024.112416","DOIUrl":"10.1016/j.asoc.2024.112416","url":null,"abstract":"<div><div>Learning Automaton (LA), a branch of reinforcement learning, initially began with the Fixed Structure Learning Automaton (FSLA) family and was later expanded to include the Variable Structure Learning Automaton (VSLA) family. Tuning the depth of the FSLA in complex environments has long been a challenging task, significantly limiting the ability to effectively navigate the exploration–exploitation dilemma. No solution has been found for this open problem yet. This study addresses this issue by introducing a novel hybrid learning automaton model called Asymmetric Variable Depth Hybrid Learning Automaton (AVDHLA). The AVDHLA model intelligently learns the depth of fixed structure LA in an autonomous manner by combining the <span><math><msub><mrow><mi>L</mi></mrow><mrow><mi>K</mi><mi>N</mi><mo>,</mo><mi>K</mi></mrow></msub></math></span> from the FSLA class and Variable Action Set LA (VASLA) from the VSLA class. Computer simulations are conducted to validate the proposed model in diverse environments, including both stationary and non-stationary (Markovian switching and State-dependent) scenarios. Performance evaluation is based on predefined metrics, such as total number of rewards (TNR) and action switching (TNAS). Statistical tests indicate that across both stationary and non-stationary environments, the AVDHLA consistently outperforms the <span><math><msub><mrow><mi>L</mi></mrow><mrow><mi>K</mi><mi>N</mi><mo>,</mo><mi>K</mi></mrow></msub></math></span> in terms of TNR and TNAS across the majority of experiments. Moreover, the AVDHLA model is applied in two key applications. Firstly, it is used to defend against the selfish mining attack in Bitcoin and is compared with the well-known tie-breaking mechanism. Simulation results consistently demonstrate that our proposed method increases the threshold for successful selfish mining attacks from 25% to 40%. Secondly, the AVDHLA model has been applied to develop a novel learning automaton-based recommendation system. The results demonstrate the superiority of the proposed method in terms of the Click-Through Rate (CTR) and Precision compared to previous approaches.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"170 ","pages":"Article 112416"},"PeriodicalIF":7.2,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143213266","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}
Pub Date : 2025-02-01DOI: 10.1016/j.asoc.2024.112645
Nima Khosravi , Hamid Reza Abdolmohammadi
Microgrids (MGs) integrate various dynamic energy sources, often making it challenging for traditional control techniques to manage these complexities. This study addresses key requirements faced by MG systems, particularly in the robust regulation of voltage and frequency (V/F) components. The proposed two-coating and layer control strategy aims to optimize MG operation effectively. To determine the coefficients for the power droop controller (PDC), a novel method called the hierarchical deep learning-based recurrent convolutional neural network (HDL-RCNN) is presented. This method incorporates feedback from system components and formats the measured data into a grid structure before inputting it into the RCNN platform. This formatting allows for the automatic extraction of temporal and spatial features crucial for V/F stability. The RCNN architecture includes several layers, such as long short-term memory (LSTM) layers, convolutional layers, highly coupled layers, and other cascaded features. Results show significant improvements in voltage and frequency control: voltage oscillations in were reduced from 0.028pu to 0.004pu, and frequency fluctuations in decreased from 0.025pu to 0.007pu. Additionally, the method ensures the voltage stabilizes at 1pu, with minimal fluctuations, and provides robust performance across dynamic load changes and noisy conditions. These findings were validated through extensive testing on a MATLAB/Simulink platform, demonstrating the effectiveness of the HDL-RCNN in enhancing V/F stability and operational reliability in MGs.
{"title":"A hierarchical deep learning-based recurrent convolutional neural network for robust voltage and frequency operation management in microgrids","authors":"Nima Khosravi , Hamid Reza Abdolmohammadi","doi":"10.1016/j.asoc.2024.112645","DOIUrl":"10.1016/j.asoc.2024.112645","url":null,"abstract":"<div><div>Microgrids (MGs) integrate various dynamic energy sources, often making it challenging for traditional control techniques to manage these complexities. This study addresses key requirements faced by MG systems, particularly in the robust regulation of voltage and frequency (V/F) components. The proposed two-coating and layer control strategy aims to optimize MG operation effectively. To determine the coefficients for the power droop controller (PDC), a novel method called the hierarchical deep learning-based recurrent convolutional neural network (HDL-RCNN) is presented. This method incorporates feedback from system components and formats the measured data into a grid structure before inputting it into the RCNN platform. This formatting allows for the automatic extraction of temporal and spatial features crucial for V/F stability. The RCNN architecture includes several layers, such as long short-term memory (LSTM) layers, convolutional layers, highly coupled layers, and other cascaded features. Results show significant improvements in voltage and frequency control: voltage oscillations in <span><math><msub><mrow><mi>MG</mi></mrow><mrow><mn>1</mn></mrow></msub></math></span> were reduced from 0.028pu to 0.004pu, and frequency fluctuations in <span><math><msub><mrow><mi>MG</mi></mrow><mrow><mn>2</mn></mrow></msub></math></span> decreased from 0.025pu to 0.007pu. Additionally, the method ensures the voltage stabilizes at 1pu, with minimal fluctuations, and provides robust performance across dynamic load changes and noisy conditions. These findings were validated through extensive testing on a MATLAB/Simulink platform, demonstrating the effectiveness of the HDL-RCNN in enhancing V/F stability and operational reliability in MGs.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"170 ","pages":"Article 112645"},"PeriodicalIF":7.2,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143213267","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}
Pub Date : 2025-02-01DOI: 10.1016/j.asoc.2024.112679
Pengrui Wang , Linfu Xie , Xiaoqiong Qin , Rong Liu
Hyperspectral remote sensing image unmixing poses a significant challenge, especially in the precise identification of pure spectral signatures (endmembers). This identification is a multifaceted optimization problem, often leading to locally optimal solutions given the inherent complexity and high dimensionality of hyperspectral images. A single objective function proves insufficient to model the rich spectral variability of endmembers. Recent advances in multiobjective evolutionary optimization for endmember bundle extraction (EBE) have illuminated the potential of multiobjective optimization techniques to address spectral variability. Yet, issues like repeated endmembers and high dimensionality have not been adequately addressed in the evolutionary process. This article introduces the Multiobjective Multi-Space Collaboration (MOMSC) model, coupled with a multiobjective genetic algorithm, to bridge these gaps. Within MOMSC, a novel multiple subspace generation strategy is devised, targeting the unveiling of spectral diversity across varied feature spaces and enhancing the capture of total spectral variability. Trios of these generated subspaces are integrated to form a multiobjective EBE framework. Given the unique nature of endmembers, tailored strategies in the genetic algorithm, such as gene segment pool allocation and spatial crowding distance calculation, are proposed to counteract high dimensionality. An innovative replacement mechanism is also proposed, refining the search space for subsequent subspace groups, circumventing the repeated endmember dilemma, and bolstering the odds of acquiring more endmember variabilities. Experimental results from three real hyperspectral images validate the effectiveness of our proposed MOMSC approach.
{"title":"Multiobjective multi-space collaboration model for addressing spectral variability in hyperspectral image unmixing","authors":"Pengrui Wang , Linfu Xie , Xiaoqiong Qin , Rong Liu","doi":"10.1016/j.asoc.2024.112679","DOIUrl":"10.1016/j.asoc.2024.112679","url":null,"abstract":"<div><div>Hyperspectral remote sensing image unmixing poses a significant challenge, especially in the precise identification of pure spectral signatures (endmembers). This identification is a multifaceted optimization problem, often leading to locally optimal solutions given the inherent complexity and high dimensionality of hyperspectral images. A single objective function proves insufficient to model the rich spectral variability of endmembers. Recent advances in multiobjective evolutionary optimization for endmember bundle extraction (EBE) have illuminated the potential of multiobjective optimization techniques to address spectral variability. Yet, issues like repeated endmembers and high dimensionality have not been adequately addressed in the evolutionary process. This article introduces the Multiobjective Multi-Space Collaboration (MOMSC) model, coupled with a multiobjective genetic algorithm, to bridge these gaps. Within MOMSC, a novel multiple subspace generation strategy is devised, targeting the unveiling of spectral diversity across varied feature spaces and enhancing the capture of total spectral variability. Trios of these generated subspaces are integrated to form a multiobjective EBE framework. Given the unique nature of endmembers, tailored strategies in the genetic algorithm, such as gene segment pool allocation and spatial crowding distance calculation, are proposed to counteract high dimensionality. An innovative replacement mechanism is also proposed, refining the search space for subsequent subspace groups, circumventing the repeated endmember dilemma, and bolstering the odds of acquiring more endmember variabilities. Experimental results from three real hyperspectral images validate the effectiveness of our proposed MOMSC approach.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"170 ","pages":"Article 112679"},"PeriodicalIF":7.2,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143213274","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}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号