Applied Soft Computing最新文献

{"title":"A multi-strategy fruit fly optimization algorithm for the distributed permutation flowshop scheduling problem with sequence-dependent setup times","authors":"Cai Zhao ,&nbsp;Lianghong Wu","doi":"10.1016/j.asoc.2024.112436","DOIUrl":"10.1016/j.asoc.2024.112436","url":null,"abstract":"<div><div>Distributed manufacturing has become one of the mainstream manufacturing modes today and is widely present in industries such as aviation and electronics. However, in actual production processes, unexpected situations such as machine failures and tool changes may occur, which require time. Based on practical needs, this paper studies a distributed permutation flow shop scheduling problem with sequence-dependent setup times (DPFSP/SDST) aimed at minimizing the makespan and proposes a hybrid multi-strategy fruit fly optimization algorithm (HMFOA) to solve it. In HMFOA, three strategies are constructed to initialize the positions of some individual flies in the solution space to improve population diversity. In the smell search phase, four problem-oriented neighborhood perturbation operators are designed, and sinusoidal optimization algorithm is introduced to control the search range, which improves the global search ability of the algorithm. In the visual search phase, a position reconstruction strategy is proposed to divide individual flies into different populations based on their mass. Through the interaction of individuals from different populations, the convergence is accelerated and the algorithm efficiency is improved. In addition, a local search strategy is designed to guide the flies to more promising areas. Based on well-known examples of DPFSP in the literature, a comprehensive test set was generated for DPFSP/SDST, taking into account various combinations of jobs, machines, factories, and SDST, resulting in 270 benchmark instances used to validate the performance of HMFOA, and compared to eight other advanced algorithms. The relative percentage deviation of HMFOA is 1.00%, which is significant improvement.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"167 ","pages":"Article 112436"},"PeriodicalIF":7.2,"publicationDate":"2024-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142653651","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 sparse diverse-branch large kernel convolutional neural network for human activity recognition using wearables","authors":"Minghui Yao ,&nbsp;Dongzhou Cheng ,&nbsp;Lei Zhang ,&nbsp;LiangDong Liu ,&nbsp;Shuangteng Song ,&nbsp;Hao Wu ,&nbsp;Aiguo Song","doi":"10.1016/j.asoc.2024.112444","DOIUrl":"10.1016/j.asoc.2024.112444","url":null,"abstract":"<div><div>During the past decade, large convolutional kernels have long been under the shadow of small convolutional kernels since the introduction of VGG backbone network. It always remains mysterious whether one can design pure convolutional neural network (CNN) while plugging larger kernels to model long-range dependency for human activity recognition (HAR), which has been rarely explored in previous literatures. In this paper, we revive the usage of larger kernels in the context of HAR and attempt to eliminate the performance gap between large kernels and small kernels by strategically applying a large receptive field, without incurring high memory and computational footprints. Built on two recipes, i.e., Diverse-Branch and Dynamic Sparsity, we design a pure CNN architecture named SLK-Net for activity recognition, which is equipped with sparse diverse-branch larger kernels. To validate the effectiveness of our approach, we perform a series of extensive experiments on four public benchmarks including UCI-HAR, WISDM, UniMiB-SHAR and USC-HAD, which show that large kernels can benefit its ability to capture long-range dependency and consistently beat state-of-the-art small-kernel counterparts across a wide range of activity classification tasks. Real activity inference latency is measured on a mobile device, which reveals that such sparse diverse-branch kernels can lead to inference speedup than vanilla large kernels. We hope this work may further inspire relevant CNN-based studies in the HAR community.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"167 ","pages":"Article 112444"},"PeriodicalIF":7.2,"publicationDate":"2024-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142654119","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 reinforcement learning hyper-heuristic algorithm for the distributed flowshops scheduling problem under consideration of emergency order insertion","authors":"Fuqing Zhao ,&nbsp;Yuebao Liu ,&nbsp;Tianpeng Xu ,&nbsp;Jonrinaldi","doi":"10.1016/j.asoc.2024.112461","DOIUrl":"10.1016/j.asoc.2024.112461","url":null,"abstract":"<div><div>Large enterprises are composed of several subproduction centers. The production plan is changed based on the procedure of the manufacturing system. The distributed flowshop scheduling problem under consideration of emergence order insertion is challenging as the assignment of the product, and the scheduling of the process are coupled with each other. A general distributed flow shop scheduling problem regarding emergency order insertion (DFSP_EOI) is addressed under rescheduling circumstances in this paper. The Q-learning hyper-heuristic algorithm with dynamic insertion rule (QLHH_DIR) is proposed to solve the DFSP_EOI. Eight low-level heuristics (LLHs) for static job assignment. A dynamic insertion rule based on the state of each production center is designed for emergency order insertion. The Q-learning mechanism at high-level space selects appropriate LLH through learning the experience from the optimization process. The computational simulation is carried out, and the results confirm that the proposed algorithm is superior to the competitors in solving the distributed flow shop rescheduling problem. The results of the 720 problem instances show that the proposed algorithm is highly efficient in rescheduling problems.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"167 ","pages":"Article 112461"},"PeriodicalIF":7.2,"publicationDate":"2024-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142654035","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":"Differential evolution with multi-strategies for UAV trajectory planning and point cloud registration","authors":"Guozhang Zhang ,&nbsp;Shengwei Fu ,&nbsp;Ke Li ,&nbsp;Haisong Huang","doi":"10.1016/j.asoc.2024.112466","DOIUrl":"10.1016/j.asoc.2024.112466","url":null,"abstract":"<div><div>The present study introduces a novel adaptive algorithm, MELSHADE-cnEpSin, which aims to enhance the performance of LSHADE-cnEpSin, which is not only stands out as one of the most competitive versions of differential evolution but also holds the distinction of being one of the CEC winner algorithms. Compared to the original methodology, three main distinctions are presented. To begin with, we adopt an adaptive selection mechanism (ASM) of crossover rate Cr value based on the external archive to rechoose a suitable value. In the next place, a nonlinear population reduction strategy using Sigmoid function is employed to improve population distribution. Additionally, a restart strategy is implemented to mitigate the risk of algorithmic convergence towards suboptimal solutions. Furthermore, the performance of MELSHADE-cnEpSin was evaluated using standard CEC2017 and CEC2022 test suites in conjunction with nine CEC-winning algorithms (L-SHADE, EBOwithCMAR, AGSK, LSHADE-SPACMA, LSHADE-cnEpSin, ELSHADE-SPACMA, EA4eig, MadDE and APGSK-IMODE) as well as two novel algorithms (ACD-DE and MIDE). Furthermore, MELSHADE-cnEpSin was effectively employed to address the challenge of UAV trajectory planning in intricate mountainous terrain and underwent simulation with point cloud registration cases utilizing a rapid global registration dataset, thereby showcasing the potential of MELSHADE-cnEpSin in tackling real-world optimization problems.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"167 ","pages":"Article 112466"},"PeriodicalIF":7.2,"publicationDate":"2024-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142654120","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":"Broad Distributed Game Learning for intelligent classification in rolling bearing fault diagnosis","authors":"Haoran Liu ,&nbsp;Haiyang Pan ,&nbsp;Jinde Zheng ,&nbsp;Jinyu Tong ,&nbsp;Mengling Zhu","doi":"10.1016/j.asoc.2024.112470","DOIUrl":"10.1016/j.asoc.2024.112470","url":null,"abstract":"<div><div>As a new Single Layer Feedforward Network (SLFN) architecture, Broad Learning System (BLS) has been widely used in the field of fault diagnosis because of its fast-training speed and high generalization capability. However, when features in different classes of signals are similar or weak, BLS generates a large number of redundant features that may be difficult to classify accurately. In view of this, a new Broad Distributed Game Learning (BDGL) method is proposed in this paper, which maps data into the game space by constructing two non-parallel game hyperplanes to achieve game and segmentation of different similar features, thereby making the data linearly differentiable in the game space. Meanwhile, a linear distribution constraint term is designed to reduce noise fitting and weak feature learning in training data learning by limiting the complexity of model parameters, thereby making the solution of the objective function simpler and faster. By comparing the Precision, Recall, F-score, Kappa and Accuracy of BDGL and the comparison methods on the two types of rolling bearing experimental data, the results show that BDGL has a high classification accuracy. In addition, the experimental results on small and noisy samples once again demonstrate the effectiveness of BDGL, which provides an efficient solution for rolling bearing fault diagnosis.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"167 ","pages":"Article 112470"},"PeriodicalIF":7.2,"publicationDate":"2024-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142654117","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":"Differential neural network based adaptive average output feedback control design for dosage determination on cancer based immunotherapy treatment","authors":"N. Aguilar-Blas ,&nbsp;I. Chairez ,&nbsp;A. Cabrera","doi":"10.1016/j.asoc.2024.112368","DOIUrl":"10.1016/j.asoc.2024.112368","url":null,"abstract":"<div><div>Immunotherapy involves natural and synthetic substances to stimulate the body’s immune response. This treatment approach is practical not only for addressing immune deficiencies but also for combating malignancies. This paper describes a non-parametric approximated adaptive control process for managing cancer dynamics under immunotherapy treatment, utilizing a combination of a differential neural network (DNN) observer and nonlinear control techniques such as sliding mode and local optimal strategies. By employing the state estimation and control methods, close tracking between the estimated states provided by the neural network and the cancer model dynamics is possible. Internal model reconstruction and an observer provided by a variable structure model are essential for controlling unknown plants. Furthermore, the control design has successfully reduced tumor cells despite uncertainties and external perturbations affecting cancer dynamics. This robustness enhances the reliability of the proposed design. A virtual real-time scheme was developed to demonstrate this controller’s feasibility in real clinical scenarios. In this scheme, a simulated patient generates variables of immunotherapy dynamics as electrical signals, which are then analyzed by a real-time project.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"167 ","pages":"Article 112368"},"PeriodicalIF":7.2,"publicationDate":"2024-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142654118","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":"Shapelet selection for time series classification","authors":"Cun Ji,&nbsp;Yanxuan Wei,&nbsp;Xiangwei Zheng","doi":"10.1016/j.asoc.2024.112431","DOIUrl":"10.1016/j.asoc.2024.112431","url":null,"abstract":"<div><div>In recent times, increasing attention has been given to shapelet-based methods for time series classification. However, in the majority of current methods, similar subsequences were often selected as shapelets, thereby reducing the final interpretability of these methods. Aiming to circumvent the selection of similar subsequences as the final shapelets, a novel shapelet selection method (SSM) was proposed in this paper. Firstly, shapelet candidates were generated by SSM through time series segmentation to avoid excessive generation of similar candidates from a single time series. Secondly, all shapelet candidates were evaluated simultaneously to improve evaluation efficiency. Finally, SSM introduced a position-based filter to prevent the selection of similar sequences repeatedly. The results obtained on the UCR TSC archive demonstrated the effectiveness of the proposed method.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"167 ","pages":"Article 112431"},"PeriodicalIF":7.2,"publicationDate":"2024-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142654033","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 fuzzy-based frame transformation to mitigate the impact of adversarial attacks in deep learning-based real-time video surveillance systems","authors":"Sheikh Burhan Ul Haque","doi":"10.1016/j.asoc.2024.112440","DOIUrl":"10.1016/j.asoc.2024.112440","url":null,"abstract":"<div><div>Deep learning (DL) techniques have become integral to smart city projects, including video surveillance systems (VSS). These advanced technologies offer significant benefits, such as enhanced accuracy and efficiency in monitoring and managing urban environments. However, despite their advantages, these systems are not without vulnerabilities. One of the most pressing challenges is their susceptibility to adversarial attacks, which can lead to critical misclassifications during inference. To address these challenges, our research focuses on developing a more robust smart city VSS. Our research unfolds across two pivotal initiatives. In our initial exploration, we introduce a pioneering framework that extends the reach of adversarial attacks to real-time VSS. A practical manifestation involved implementing a real-time face mask surveillance system based on Multi-Task Cascaded Convolutional Networks (MTCNN) for face detection and MobileNet-v2 for face mask classification, subjecting it to the Fast Gradient Sign Method (FGSM) adversarial attack in real-time. In our subsequent endeavor, we propose a sophisticated defense mechanism deploying Fuzzy Image Transformation as a pre-processing unit (FITP). This strategic defense fortification significantly reinforces our real-time VSS against adversarial intrusions. Experimental findings highlight the effectiveness of the proposed adversarial attack framework in real-time, resulting in a marked reduction in the model's performance from a precision (P) of 93 %, recall (R) of 93 %, F1 score (F) of 93 %, and accuracy (A) of 93–22 %, 21 %, 22 %, and 22 %, respectively. However, the post-implementation efficacy of our defense mechanism is striking, enhancing the model's average performance to a noteworthy improvement, with P, R, F, and A ascending to 91 %, 90 %, 91 %, and 91 %. This research illuminates the vulnerabilities intrinsic to VSS in the face of adversarial threats, underscoring the critical need for heightened awareness and the development of robust defense mechanisms before real-world deployment.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"167 ","pages":"Article 112440"},"PeriodicalIF":7.2,"publicationDate":"2024-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142654116","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":"Cross-functional group decision making with heterogeneous cooperation for digital transformation in supply chain resilience","authors":"Ming Tang ,&nbsp;Huchang Liao","doi":"10.1016/j.asoc.2024.112463","DOIUrl":"10.1016/j.asoc.2024.112463","url":null,"abstract":"<div><div>Supply chain resilience plays a critical role in gaining competitive advantages for companies. The resilience of supply chains can be achieved by leveraging emerging digital technologies to realize digital transformation. It is necessary to select an appropriate digitalization technology under such background. The wide-spanning of digital transformation and technology selection needs cross-functional integration of various expertise. However, in the process of making decisions by leveraging expert wisdom, differences in experts’ willingness to cooperate lead to difficulties in reaching a consensus. The existing literature fails to incorporate both non-cooperation and proactive-cooperation into the consensus reaching process. Thus, in this study, we introduce a cross-functional multi-attribute group decision making model for digitalization technology selection. To manage potential non-cooperative behaviors in the group consensus reaching process, the proposed model allows experts to have proactive cooperation, i.e., making more contributions than recommended feedback suggestions provided by the moderator. Proactive cooperation can make up for the loss caused by the non-cooperative behaviors of experts. A knowledge mining method is proposed to mine academic and practical preferences for attributes. Two consensus mechanisms are put forward for the meso decision-making process in functional teams and the macro decision-making process in the whole group, respectively. An illustrative example regarding the technology selection in shipbuilding industry is provided to verify the applicability of our model. Numerical experiments suggest that our model will improve the efficiency of consensus reaching process.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"167 ","pages":"Article 112463"},"PeriodicalIF":7.2,"publicationDate":"2024-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142654038","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":"Document-level multiple relations extraction method via evidence guidance and relation correlation","authors":"Hao Yang ,&nbsp;Qiming Fu ,&nbsp;You Lu ,&nbsp;Yunzhe Wang ,&nbsp;Lanhui Liu ,&nbsp;Jianping Chen","doi":"10.1016/j.asoc.2024.112391","DOIUrl":"10.1016/j.asoc.2024.112391","url":null,"abstract":"<div><div>Document-level Relation Extraction (DocRE) aims to extract semantic relations between entity pairs, spanning multiple sentences, paragraphs or even the entire document. These relations can often be predicted by partial sentences within the document, the evidence sentence. However, the relation derived only from sentence information is incomplete, because it ignores the case of multiple relations between entity pairs. Therefore, how to select effective evidence sentences and how to predict multiple relations more accurately have become challenges for the existing DocRE models. In response to these challenges, we introduce Reinforcement Learning (RL) to select more effective evidence sentences, while using heuristic rules to narrow down the search space of RL. Secondly, we utilize GAT to acquire the features of co-occurrence relations, which can greatly improve multiple relations prediction performance. Moreover, the combination of the features of co-occurrence relations and the evidence sentence information enables our method to achieve both high effectiveness and precision. The experimental results show that, compared with other advanced methods, our method achieves an <span><math><msub><mrow><mi>F</mi></mrow><mrow><mn>1</mn></mrow></msub></math></span> score of 66.56 and the <span><math><mrow><mi>E</mi><mi>v</mi><mi>i</mi></mrow></math></span> <span><math><msub><mrow><mi>F</mi></mrow><mrow><mn>1</mn></mrow></msub></math></span> score of 56.69, which attains the state-of-the-art performance on public datasets.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"167 ","pages":"Article 112391"},"PeriodicalIF":7.2,"publicationDate":"2024-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142654036","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}