Applied Soft Computing最新文献

{"title":"Short-term forecasting of electricity price using ensemble deep kernel based random vector functional link network","authors":"Someswari Perla ,&nbsp;Ranjeeta Bisoi ,&nbsp;P.K. Dash ,&nbsp;A.K. Rout","doi":"10.1016/j.asoc.2025.113012","DOIUrl":"10.1016/j.asoc.2025.113012","url":null,"abstract":"<div><div>Accurate short-term electricity price forecasting in a deregulated electrical market is a difficult task as the electricity price exhibits high nonlinearity, sharp price spikes, and seasonality in different frequencies, etc. Thus, this study presents a new approach using an Ensemble Deep Kernel Random Vector Functional Link Network (EDKRVFLN) model hybridized with a Chaotic Sine Cosine Improved Firefly Algorithm (CSCIFA) for short-term electricity price forecasting with better generalization capacity, simple structure, and significant accuracy. Unlike the Ensemble Deep Random Vector Functional Link Network (EDRVFLN) where each stacked layer requires proper choice of the number of hidden nodes and manual tuning of random weights and biases along with the pseudoinverse solution of the output weights in each layer leading to suboptimal model generalization. However, the choice of random weights and biases along with the number of hidden neurons in the proposed EDKRVFLN model can be dispensed by using kernel-based transformation and representation learning. Further each stacked layer of the proposed model utilizes kernel based linear features from the direct links and nonlinearly transformed features from the enhancement nodes from the preceding layers of the prediction model. Also, each layer produces an output by simple invertible kernel matrix inversion based on generalized least squares, and the final output is the ensemble of the outputs from each layer, thus simultaneously producing an ensemble and deep learning framework. Seven electricity price datasets are examined to confirm the supremacy of the proposed model in comparison to several benchmark models.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"174 ","pages":"Article 113012"},"PeriodicalIF":7.2,"publicationDate":"2025-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143686926","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":"Series clustering and dynamic periodic patching-based transformer for multivariate time series forecasting","authors":"Yijie Wang ,&nbsp;Xiao Wu ,&nbsp;Jiaying Zhang ,&nbsp;Weiping Wang ,&nbsp;Linjiang Zheng ,&nbsp;Jiaxing Shang","doi":"10.1016/j.asoc.2025.112980","DOIUrl":"10.1016/j.asoc.2025.112980","url":null,"abstract":"<div><div>Multivariate time series forecasting (MTSF) is widely employed in research-intensive domains, such as weather forecasting. Recently, Transformer-based models have outstanding ability to achieve SOTA performance, benefiting from its self-attention mechanism. However, existing models fall short in capturing multivariate inter-dependencies and local semantic representations. To tackle the above limitations, we propose a series clustering and dynamic periodic patching-based Transformer model named CMDPPformer, with two distinctive characteristics: (1) A channel-mixing module based on series clustering is proposed which can strengthen the association between variables with high sequence similarity, and weaken the effect of uncorrelated variables. Concretely, we use whole-time series clustering to group multivariate time series into clusters. After that, variables in the same cluster share the same Transformer backbone while variables in different clusters do not affect each other. (2) A dynamic periodic patching module is introduced which can better capture semantic information and improve Transformer’s local semantic representation. Concretely, multivariate time series after clustering are dynamically segmented into periodic patches as Transformer’s input token. Experimental results show that CMDPPformer can achieve an overall 13.76% and 10.16% relative improvements than SOTA Transformer-based models on seven benchmarks, covering four real-world applications: energy, weather, illness and economic.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"174 ","pages":"Article 112980"},"PeriodicalIF":7.2,"publicationDate":"2025-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143629469","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":"Enhancing AI safety of machine unlearning for ensembled models","authors":"Huanyi Ye ,&nbsp;Jiale Guo ,&nbsp;Ziyao Liu ,&nbsp;Yu Jiang ,&nbsp;Kwok-Yan Lam","doi":"10.1016/j.asoc.2025.113011","DOIUrl":"10.1016/j.asoc.2025.113011","url":null,"abstract":"<div><div>Recently, machine unlearning (MU) has received significant attention for its ability to remove specific undesired knowledge from a trained model, thereby ensuring AI safety. Furthermore, efforts have been made to integrate MU into existing Machine Learning as a Service (MLaaS), allowing users to raise requests to remove the influence of their data used in the training phase, after which the server conducts MU to remove its influence based on the unlearning requests. However, previous research reveals that malicious users may manipulate the requests so that the model utility may be significantly compromised after unlearning, which is known as malicious unlearning. In addition, privacy leakage may be exploited by malicious users by analyzing inference results obtained from the original model and the unlearned model. In this connection, we investigate these potential risks, specifically in ensemble models, which are widely adopted in MU because of their efficiency in unlearning and robustness in learning. However, despite these advantages, their vulnerabilities to malicious unlearning and privacy leakage remain largely unexplored. Our work explores malicious unlearning and malicious inference in ensemble settings. We propose a method in which malicious unlearning requests can trigger hidden poisons in ensembles, causing target images to be misclassified as intended by adversaries. Additionally, we introduce a privacy leakage attack where adversaries with black-box access to voting outputs can infer the unlearned label by analyzing the differences between the original and unlearned ensemble outputs. Experimental results demonstrate that these attacks can be highly stealthy and achieve a high success rate. Furthermore, comparative experiments reveal that these attacks present slightly lower stealthiness in ensemble settings compared to single-model scenarios, suggesting that ensemble models have advantages in detecting such malicious activities. These findings reveal that ensemble models are vulnerable to malicious unlearning and privacy leakage and highlight the urgent need for more robust MU designs to ensure AI safety.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"174 ","pages":"Article 113011"},"PeriodicalIF":7.2,"publicationDate":"2025-03-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143687354","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":"Discrete Differentiated Creative Search for traveling salesman problem","authors":"Qi Xu,&nbsp;Kewen Xia,&nbsp;Xiaoyu Chu","doi":"10.1016/j.asoc.2025.112998","DOIUrl":"10.1016/j.asoc.2025.112998","url":null,"abstract":"<div><div>A novel population-based Discrete Differentiated Creative Search (DDCS) is proposed in this paper for solving the traveling salesman problem (TSP). DDCS introduces greedy beam search to adaptively initialize the population and improve the quality of the initial solutions. Second, a multi-edge construction operator, edge-based mathematical operations and a similarity attraction operator are used to guide individuals from different population categories towards higher-quality solutions based on the current solutions. Finally, a random nearest neighbor replacement strategy is used to replace individuals with the same distance heuristically, reducing the assimilation rate of the population. DDCS is tested with 50 instances from TSPLIB and compared with a variety of state-of-the-art and variants of classical algorithms. The results demonstrate that DDCS exhibits superior optimization capability and higher stability.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"174 ","pages":"Article 112998"},"PeriodicalIF":7.2,"publicationDate":"2025-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143686928","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":"Adaptive encoding and comprehensive attention decoding network for medical image segmentation","authors":"Xin Shu ,&nbsp;Aoping Zhang ,&nbsp;Zhaoyang Xu ,&nbsp;Feng Zhu ,&nbsp;Wei Hua","doi":"10.1016/j.asoc.2025.112990","DOIUrl":"10.1016/j.asoc.2025.112990","url":null,"abstract":"<div><div>Medical image segmentation involves partitioning different tissues or lesion areas within medical images. Achieving automatic segmentation can markedly improve efficiency and accuracy, which is significant for biomedical clinical diagnosis. With the rapid development of deep convolutional neural networks (DCNN), U-Net has been widely used in medical image segmentation due to its encoder-decoder structure and skip connection. However, it is still hard for U-Net to handle certain challenging cases. In this study, we propose an adaptive encoding and comprehensive attention decoding network (AA-Net), which is derived from U-Net to address the issues of the semantic gap as well as the loss of spatial information during convolutions. AA-Net takes into account the different characteristics of the encoder and decoder. In the encoder, we design a simple Adaptive Calibration Module (ACM) to improve the representation ability of candidate features. In the decoder, we introduce a Comprehensive Attention Feature Extraction (CAFE) module, which employs multiple attention mechanisms after feature fusion to alleviate the semantic gap. Benefiting from CAFE, AA-Net can better handle the challenging cases where the segmentation targets vary in position, size, and scale. Additionally, we suggest a weighted hybrid loss function for precise boundary segmentation. We validate the effectiveness of AA-Net and each component on three biomedical image datasets. The results demonstrate that our method outperforms state-of-the-art methods in different medical segmentation tasks, proving it is lightweight, efficient, and general.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"174 ","pages":"Article 112990"},"PeriodicalIF":7.2,"publicationDate":"2025-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143637196","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":"Geometrical invariant generative invisible hyperlinks based on feature points","authors":"Zecheng Peng ,&nbsp;Bingwen Feng ,&nbsp;Xiaotao Xu ,&nbsp;Jilian Zhang ,&nbsp;Donghong Cai ,&nbsp;Wei Lu","doi":"10.1016/j.asoc.2025.112959","DOIUrl":"10.1016/j.asoc.2025.112959","url":null,"abstract":"<div><div>To enhance the visual diversity of Quick Response (QR) codes while ensuring their robust decoding capabilities, this paper introduces an innovative invisible hyperlink generation system. The system can use a message sequence to directly generate a hyperlink image. By harnessing the latent space of a suggested feature point generation network, the system extends the robustness of image feature points to the hyperlink images it generates. Specially, an image generation network is first designed to synthesize high-quality images based on feature point data. Subsequently, a set of lightweight message encoder and decoder are introduced to embed message bits into the latent space of the image generation network. Experimental results show that the proposed invisible hyperlink generation system can successfully generate images containing hyperlinks, exhibiting remarkable resilience against common signal processing and geometric distortions. It harbors diverse potential applications, encompassing website URLs, contact information, product specifics, and numerous other use cases.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"174 ","pages":"Article 112959"},"PeriodicalIF":7.2,"publicationDate":"2025-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143629920","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":"Adaptive large neighborhood search incorporating mixed-integer linear programming for electric vehicle routing problem with mobile charging and nonlinear battery degradation","authors":"Senyan Yang ,&nbsp;Ruiyan Zhang ,&nbsp;Ying Ma ,&nbsp;Xingquan Zuo","doi":"10.1016/j.asoc.2025.112988","DOIUrl":"10.1016/j.asoc.2025.112988","url":null,"abstract":"<div><div>The limited driving range and short battery life are obstacles to the widespread adoption of electric vehicles in urban logistics. This study proposes an electric vehicle routing problem with time window, mobile charging, and nonlinear battery degradation. Mobile charging vehicles (MCVs) can be flexibly scheduled to charge the electric delivery vehicles (EDVs) at customer locations, reducing the electricity consumption caused by the detours to the charging stations. The proposed problem is formulated into an arc-based model that incorporates nonlinear battery degradation costs associated with State of Charge (SOC) and charging strategies, thereby enhancing the complexity of the spatio-temporal synchronization mechanism. Constraining a lower SOC can mitigate the battery degradation of EDVs, but it leads to increased charging demands and makes searching for feasible routing solutions more challenging due to the interdependence between MCVs and EDVs. A hybrid adaptive large neighborhood search heuristic algorithm is developed. Dynamic programming is embedded in the algorithm framework to devise charging schemes considering nonlinear battery degradation for the given EDVs’ routes. A mixed-integer linear programming model is formulated to select the combination of labels with continuous charging decisions and design MCVs’ routes. Extensive numerical experiments are conducted to verify the proposed model and algorithm. Experimental results indicate considering battery degradation in the objectives significantly improves the total system costs by optimizing the SOC and charging quantity. Mobile charging can be an alternative for constructing fixed charging facilities due to the charging flexibility of MCVs. The performance of our algorithm is demonstrated through both large-scale instances and a real-world case study on urban logistics.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"175 ","pages":"Article 112988"},"PeriodicalIF":7.2,"publicationDate":"2025-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143724230","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":"Multi-agent modeling for indoor fire risk prediction during evacuation based on cellular automata and artificial neural network","authors":"Peng Lu","doi":"10.1016/j.asoc.2025.113013","DOIUrl":"10.1016/j.asoc.2025.113013","url":null,"abstract":"<div><div>Fire cases have always posed threats to human lives and property safety, and new approaches have been developed to investigate how people behave during the fire process. Understanding the underlying mechanism under specific scenarios and conditions is critical to find possible ways of reducing social losses. Here, we propose a coupled model that combines FDS and CA, to assess fire risks in a multi-story dormitory building at a university. For this real target case, the settings of automatic sprinklers and temperature alarms will be considered in our coupled model. The aim is to investigate how pedestrians behave under the fire emergencies and how fire safety facilities (exits) shape final evacuation outcomes. To analysis the final outcomes and related factors, we use Event Tree and BP neural network methods to assess and predict individual risk levels. It suggests that controlling the number of people in each dormitory will effectively reduce the fire risk, and the existence of safety facilities can significantly contain fire risks. Early fire warning systems and quick response times are critical to reduce casualties during the evacuation process. Individual risk levels can be efficiently calculated by Event Tree method, and BP neural network can accurately predict fire risk levels. By integrating technologies such as FDS, CA, ETA, and BP neural networks, our model can effectively simulate the dynamic process of the fire evacuation while accurately predicting the fire risks, which establishes an effective link between environmental factors and fire risk assessment. This provides a methodological reference for future fire risk assessment research.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"174 ","pages":"Article 113013"},"PeriodicalIF":7.2,"publicationDate":"2025-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143686925","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":"Multi-agent reinforcement learning system framework based on topological networks in Fourier space","authors":"Licheng Sun,&nbsp;Ao Ding,&nbsp;Hongbin Ma","doi":"10.1016/j.asoc.2025.112986","DOIUrl":"10.1016/j.asoc.2025.112986","url":null,"abstract":"<div><div>Currently, multi-agent reinforcement learning (MARL) has been applied to various domains such as communications, network management, power systems, and autonomous driving, showcasing broad application scenarios and significant research potential. However, in complex decision-making environments, agents that rely solely on temporal value functions often struggle to capture and extract hidden features and dependencies within long sequences in multi-agent settings. Each agent’s decisions are influenced by a sequence of prior states and actions, leading to complex spatiotemporal dependencies that are challenging to analyze directly in the time domain. Addressing these challenges requires a paradigm shift to analyze such dependencies from a novel perspective. To this end, we propose a Multi-Agent Reinforcement Learning system framework based on Fourier Topological Space from the foundational level. This method involves transforming each agent’s value function into the frequency domain for analysis. Additionally, we design a lightweight weight calculation method based on historical topological relationships in the Fourier topological space. This addresses issues of instability and poor reproducibility in attention weights, along with various other interpretability challenges. The effectiveness of this method is validated through experiments in complex environments such as the StarCraft Multi-Agent Challenge (SMAC) and Google Football. Furthermore, in the Non-monotonic Matrix Game, our method successfully overcame the limitations of non-monotonicity, further proving its wide applicability and superiority. On the application level, the proposed algorithm is also applicable to various multi-agent system domains, such as robotics and factory robotic arm control. The algorithm can control each joint in a coordinated manner to accomplish tasks such as enabling a robot to stand upright or controlling the movements of robotic arms.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"174 ","pages":"Article 112986"},"PeriodicalIF":7.2,"publicationDate":"2025-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143637200","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 three-way decision-based model for occupational risk assessment and classification in the healthcare industry","authors":"Ran Liu ,&nbsp;Hu-Chen Liu ,&nbsp;Qi-Zhen Zhang ,&nbsp;Hua Shi","doi":"10.1016/j.asoc.2025.112991","DOIUrl":"10.1016/j.asoc.2025.112991","url":null,"abstract":"<div><div>Nowadays, occupational health and safety risk assessment (OHSRA) has gained more importance since occupational hazards can cause loss of life, injuries, delays, and cost overruns in an organization. The OHSRA is a critical activity for identifying, analyzing and reducing the potential occupational hazards arising from workplace for corrective actions. In this study, a new OHSRA model is proposed for the risk assessment and classification of occupational hazards by utilizing the criteria importance through inter-criteria correlation (CRITIC) method and three-way decision (TWD). First, the 2-tuple linguistic variables are utilized to express the complex and uncertain risk assessments of occupational hazards provided by experts. Second, an extended CRITIC method is employed to compute the weights of risk criteria by considering their interactions. Then the TWD is improved to determine the risk classifications of occupational hazards by considering their correlations. Finally, a practical case in the healthcare industry is provided to illustrate the feasibility and strengths of the proposed OHSRA model. The results show that the proposed OHSRA model can generate more credible risk classifications of occupational hazards and offer a flexible way for analyzing the risk of occupational hazards.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"174 ","pages":"Article 112991"},"PeriodicalIF":7.2,"publicationDate":"2025-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143642575","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}