Applied Soft Computing最新文献

{"title":"Discriminative distance metric learning via class-center guidance","authors":"Shijie Zhao ,&nbsp;Liang Cai ,&nbsp;Fanshuai Meng ,&nbsp;RongHua Yang","doi":"10.1016/j.asoc.2025.113552","DOIUrl":"10.1016/j.asoc.2025.113552","url":null,"abstract":"<div><div>Distance metric learning is a technique of great importance to machine learning and data processing, which can effectively improve the generalization performance of algorithms related to distance metrics. The method projects the original data to the metric space through a transformation to realize the automatic adjustment of the distance between samples, so as to achieve the increase of the between-class distance and the decrease of the within-class distance. To better achieve this goal, we propose a discriminative distance metric learning via class-center guidance (DML-CG). The proposed DML-CG learns a novel discriminative distance metric by maximizing the trace ratio of between-class covariance to within-class covariance, and at the same time transforms the trace ratio problem into a ratio-trace problem to find the global optimal solution. In addition, this method selects <em>k</em> nearest neighbors for each training sample to generate sample pairs, and jointly uses local metrics learned from multiple class-center guidance and a global metric to guide samples of the same class closer to the class center, and samples of different class farther away from the sample class center. This achieves both the distance metric and captures the discriminative structure of the data. Meanwhile, global regularization is introduced to improve the generalization performance and control overfitting. We design an alternating iteration algorithm to optimally solve the proposed method and theoretically analyze the convergence and complexity. Finally, the effectiveness of the proposed algorithm is demonstrated on structured artificial datasets and UCI datasets as well as unstructured image recognition datasets. Most of the results show that the proposed algorithm outperforms other state-of-the-art distance metric learning methods.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"182 ","pages":"Article 113552"},"PeriodicalIF":7.2,"publicationDate":"2025-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144614882","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":"AGNSA: Adaptive graph learning-based unsupervised feature selection with non-convex sparse autoencoder","authors":"Lin Sun ,&nbsp;Mengqing Li ,&nbsp;Weiping Ding ,&nbsp;Jiucheng Xu","doi":"10.1016/j.asoc.2025.113550","DOIUrl":"10.1016/j.asoc.2025.113550","url":null,"abstract":"<div><div>Some unsupervised feature selection methodologies cannot consider the two local structures for samples and features, and there are unreasonable local structures that cannot control the feature redundancy well. So, we study an adaptive graph learning-based unsupervised feature selection with a non-convex sparse autoencoder. Firstly, a single-layer autoencoder is used to construct a reconstruction loss function to reconstruct the original features, and a new Mish activation function is studied to optimize the autoencoder structure. In the autoencoder, a feature similarity matrix is established by integrating Gaussian kernel function and Euclidean distance for reflecting the similarity of features to learn the local structure of feature graph. Particularly, a non-convex regularization term is applied into a weight matrix between the input layer and hidden layer of autoencoder, and then a feature weight matrix with sparser rows can be obtained to realize feature selection. Secondly, the Gaussian kernel function and Euclidean distance are combined to establish a sample similarity matrix. In the process of auto-encoder optimization, this sample local structure is learned by updating the sample similarity matrix adaptively, and the learned local structure is constrained near the original sample similarity matrix to avoid unreasonable local structure. Then, cosine similarity is employed to consider the feature correlation and learn redundancy matrix to control the redundancy of selected features. Finally, a new objective function is constructed, and an alternating iteration scheme is designed to optimize and compute the objective function to obtain an optimal solution for parameters, where the importance of features is judged according to the obtained feature weight matrix, and the representative feature subset is selected. Experimental results illustrate this developed methodology will be better than other comparative schemes on eight high-dimensional datasets for benchmark classification.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"181 ","pages":"Article 113550"},"PeriodicalIF":7.2,"publicationDate":"2025-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144579984","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 review on metaheuristics for solving home health care routing and scheduling problems","authors":"Yaping Fu ,&nbsp;Jie Dong ,&nbsp;Kaizhou Gao ,&nbsp;Ponnuthurai Nagaratnam Suganthan ,&nbsp;Min Huang ,&nbsp;Lihua Zhu","doi":"10.1016/j.asoc.2025.113560","DOIUrl":"10.1016/j.asoc.2025.113560","url":null,"abstract":"<div><div>Nowadays, the healthcare of elderly people catches wide attention since the increase of aging population puts significant pressure on public medical resources. The population aging and scarce care resources are likely to result in a substantial increase in demand for home health care (HHC) services. In order to improve operation efficiency and reduce service cost, home health care routing and scheduling problems (HHCRSPs) have received many concerns from academia and industry in recent years. HHC provides care services to customers and elderly at their homes by assigning proper caregivers and resources as their requirements. Due to the large scale and strong coupling features, as well as various practical requirements in concrete scenarios, there exist a great many of challenges in addressing the HHCRSPs effectively. Recently, metaheuristics have been employed and made breakthroughs in solving such difficult HHCRSPs. This article aims to provide a comprehensive literature review of metaheuristics for handling HHCRSPs. The existing studies regarding HHCRSPs are firstly summarized and analyzed from the perspectives of optimization objectives, number of objectives, uncertainties, constraint conditions, and number of care centers. Next, the metaheuristic algorithms for handling the HHCRSPs are summed up and dissected from the views of classifications, solution encoding and decoding, ensemble of metaheuristics and local search strategies, test instances, performance metrics, statistical analysis, and stopping criteria. Subsequently, challenges of addressing the HHCRSPs are analyzed and discussed. Afterwards, we point out future research directions and significant research contents. Finally, we conclude this research.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"182 ","pages":"Article 113560"},"PeriodicalIF":7.2,"publicationDate":"2025-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144597271","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":"AI-fused construction portfolio investment system with risk hedging using machine learning and long-short strategies","authors":"Jui-Sheng Chou,&nbsp;Kai-Chun Lin,&nbsp;Tran-Bao-Quyen Pham","doi":"10.1016/j.asoc.2025.113555","DOIUrl":"10.1016/j.asoc.2025.113555","url":null,"abstract":"<div><div>Developing consistently profitable investment strategies presents a considerable challenge within the intricate and continuously evolving financial landscape. This manuscript introduces an automated investment model meticulously designed to optimize returns through dynamic portfolio management, with a focus on comprehensive short-term portfolio decision-making. Leveraging advanced methodologies, including machine learning, natural language processing (NLP), and deep learning techniques, this study develops a robust system capable of integrating various data sources, such as extensive financial indicators, technical analysis metrics, and sentiment analysis derived from NLP-based models. We delineate essential financial factors using extreme gradient boosting and are trained on historical transaction data, financial indices, and detailed technical indicators. Furthermore, the model incorporates transformer-based NLP techniques to extract sentiment and market insights from textual data. The system autonomously identifies optimal long-short portfolio combinations and trading opportunities, employing dynamic weight adjustments informed by predictive analytics and technical indicators. Simulation results demonstrate that dynamically weighted portfolios can effectively respond to diverse economic conditions, yielding stable returns and reduced volatility, regardless of market direction. Although the scope of this study is confined to the listed construction sector in Taiwan, backtesting substantiates the robustness and potential scalability of the proposed methodology. Future research may seek to explore broader market applications to further validate the generalizability of the approach; nonetheless, the current findings already indicate significant promise for practical implementation in medium-frequency trading strategies.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"183 ","pages":"Article 113555"},"PeriodicalIF":7.2,"publicationDate":"2025-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144703044","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":"Automated intelligent-agent optimisation of per-lane variable speed limits","authors":"Amirreza Kandiri ,&nbsp;Maria Nogal ,&nbsp;Beatriz Martinez-Pastor ,&nbsp;Rui Teixeira","doi":"10.1016/j.asoc.2025.113554","DOIUrl":"10.1016/j.asoc.2025.113554","url":null,"abstract":"<div><div>Recent advancements in intelligent transportation systems and data analytics within transportation systems present a significant opportunity to enhance operational efficiency. In this context, the pivotal role of intelligent agents in achieving real-time optimisation for traffic management is highlighted. Such agents can predict and decide autonomously and can be trained to understand the underlying complexities of the traffic in real-time. In this paper, an innovative framework to perform real-time traffic optimal management decisions is proposed. Its rationale uses a fusion of data observations and simulation to enable an autonomous agent capable of accurate adaptive traffic management. A Case Study of application is developed using the M50 motorway in Dublin, where the speed limits are applied as adaptive parameters for optimal traffic management. Results show that the intelligent agent can autonomously predict travel times and decide in real-time the optimal speed limits to impose on a motorway when signs of congestion are found. The agent can reduce the mean travel time of a time interval by up to 55 % and the mean waiting time by up to 69 % in a situation of congestion. The average travel times of the studied M50 junction have significantly improved, showing the potential of autonomous agents in enhancing real-time optimal traffic management.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"181 ","pages":"Article 113554"},"PeriodicalIF":7.2,"publicationDate":"2025-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144569799","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":"Cold chain delivery route modeling and optimizing based on the clustered Whale Optimization Algorithm","authors":"Zhe Sun,&nbsp;Shengnan Ma,&nbsp;Yongbo Jian,&nbsp;Yubin Lu,&nbsp;Zhixin Sun","doi":"10.1016/j.asoc.2025.113544","DOIUrl":"10.1016/j.asoc.2025.113544","url":null,"abstract":"<div><div>In addressing the optimization problem of cold chain logistics distribution paths under multiple constraints, comprehensive consideration is given to refrigeration parameters, cargo damage rates, and carbon emission factors. Systematic analysis is performed to quantify the combined effects of load capacity and ambient temperature on total operational costs. A traffic condition monitoring mechanism is subsequently integrated to dynamically evaluate roadway statuses, thereby enabling the acquisition of empirically validated transportation durations. Based on these operational parameters, a traffic-responsive optimization model for cold chain logistics (CCL) distribution routes is formulated. To address the complex multimodal characteristics of the model, the Clustering Whale Optimization Algorithm (CWOA) is proposed. Density-Based Spatial Clustering of Applications with Noise (DBSCAN) clustering is employed to achieve dynamic population reorganization. An innovative path encoding rule based on search agent addresses is developed, with a sine-cosine oscillation operator introduced to replace linear search strategies during stochastic search processes, thereby enhancing the flexibility of individual search movements. Comparative testing on 23 benchmark functions from the IEEE Congress on Evolutionary Computation (CEC) effectively verifies CWOA's high precision and rapid convergence performance. The model and algorithm are subsequently applied to simulation experiments for cold chain logistics distribution in the Yangtze River Delta region, demonstrating CWOA's superior capability in solving CCL distribution path planning problems.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"182 ","pages":"Article 113544"},"PeriodicalIF":7.2,"publicationDate":"2025-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144597268","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 grid-based MOEA guided by synchronous self-adaptive local diversity-preserving for constrained optimization of an extended-range small self-defense missile","authors":"Hao Yan,&nbsp;Xiaobing Zhang","doi":"10.1016/j.asoc.2025.113511","DOIUrl":"10.1016/j.asoc.2025.113511","url":null,"abstract":"<div><div>Miniaturized airborne antimissile interception systems can increase a fighter’s payload capacity and enhance active protection capability. However, conventional small missiles have a large aspect ratio, which reduces propulsion system reliability and aerodynamic maneuverability. This study proposes a two-stage separable extended-range missile design to solve this problem. The multidisciplinary design optimization (MDO) model of the extended-range small self-defense missile (ERSSDM) with a nonlinear design objective space and multiple constraints produces an unwonted diversity loss problem, restricting the application of heuristic multi-objective algorithms to missile MDO problems. Therefore, two concepts-grid crowding degree and relaxation factor-are introduced, and a grid-based multi-objective evolutionary algorithm (MOEA), GMOEA-SSLD, is proposed and coupled to the MDO model of this weapon system to obtain Pareto optimal designs with well-preserved diversity. This algorithm, which uses grid-based techniques and synchronous diversity-preserving approaches, eliminates the necessity for coordinate specification and improves the design efficiency in the multidisciplinary optimization of the small missile as compared to another coordinate-based MOEA. The MDO model is decoupled to some extent based on an efficient global sensitivity analysis (GSA) approach.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"181 ","pages":"Article 113511"},"PeriodicalIF":7.2,"publicationDate":"2025-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144569955","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":"Lightweight unmanned aerial vehicles anomaly detection model based on synaptic evolution mechanism and layer-adaptive neural network","authors":"Rong Zeng ,&nbsp;Hongli Deng ,&nbsp;Bochuan Zheng ,&nbsp;Yu Lu","doi":"10.1016/j.asoc.2025.113536","DOIUrl":"10.1016/j.asoc.2025.113536","url":null,"abstract":"<div><div>The wide application of unmanned aerial vehicles (UAVs) puts strict requirements on reliable operation, and anomaly detection is a crucial method to ensure the reliability of UAVs. Existing anomaly detection models are highly dependent on time-series log data, and models based on Long Short-Term Memory (LSTM) are widely used due to their effectiveness in processing time-series data. However, the complex internal structure of LSTM involves many learning parameters. In addition, the traditional static parameter pruning methods fail to balance the conflict between performance and parameter scale dynamically. To address the above problems, this paper proposes a lightweight anomaly detection model based on the synaptic evolutionary mechanism and layer-adaptive neural network (LUV-DSA), which can be deployed in resource-constrained UAV application scenarios. Firstly, LUV-DSA simplifies the internal structure of LSTM by optimising the cell state update process with a new linearly weighted computational method. Secondly, inspired by the evolution of biological synapses, a method for intra-layer parameter pruning and inter-layer structured pruning is designed. For intra-layer parameters, LUV-DSA achieves dynamic model parameter competition by simulating the self-optimisation of synapses, minimising parameter scale while ensuring performance. For inter-layer structures, LUV-DSA enables inter-layer adaptation by calculating plasticity factors to assess the contribution of each layer. The experimental results show on seven UAV datasets that the model significantly reduces the number of parameters and inference time while ensuring accuracy. For example, on the ALFA dataset, LUV-DSA achieves 99.51 % accuracy with 96.14 % fewer parameters than MobileNetV4.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"182 ","pages":"Article 113536"},"PeriodicalIF":7.2,"publicationDate":"2025-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144597270","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":"Symbolic regression-aided hyperparameter relationship for developing ANN for fragility prediction","authors":"Mohammadreza Parvizi,&nbsp;Kiarash Nasserasadi,&nbsp;Ehsan Tafakori","doi":"10.1016/j.asoc.2025.113485","DOIUrl":"10.1016/j.asoc.2025.113485","url":null,"abstract":"<div><div>Accurate prediction of seismic fragility parameters is crucial for assessing earthquake risks and developing effective mitigation strategies. Traditional methods, such as Incremental Dynamic Analysis (IDA), impose high computational costs, limiting their practical applicability for large-scale fragility evaluations. To address this challenge, this study proposes an optimized Artificial Neural Network (ANN) architecture for predicting fragility functions of low-rise steel moment frames. Three metaheuristic optimization algorithms, including Particle Swarm Optimization (PSO), Genetic Algorithm (GA), and Bayesian Optimization (BO), were employed to optimize the number of hidden layers, the number of neurons per layer, and the learning rate of the neural network. A comparative analysis of these methods indicated that PSO outperformed the others, yielding a lower cost function value and demonstrating more excellent stability in model tuning. Additionally, the optimal learning rate in PSO was lower than in the other two methods, suggesting a slower training process but enhanced stability of the final model. Symbolic Regression (SR) was utilized to enhance prediction accuracy and derive mathematical relationships for estimating the optimal number of neurons in hidden layers using the results of optimized network architectures. As a result, based on the proposed formula, the average prediction error was reduced by approximately 23 %, demonstrating the effectiveness of the developed approach. ANN models trained based on these relationships significantly reduced computational costs while enhancing fragility prediction accuracy. Furthermore, sensitivity analysis using the Shapley Additive explanations (SHAP) algorithm was conducted to quantify the influence of input parameters on model outputs. The results indicated that structural ductility and soil type had the most significant impact on fragility estimates, whereas seismic hazard level and importance factor exhibited the least influence. These findings highlight the effectiveness of integrating ANN, metaheuristic optimization, and sensitivity analysis in developing an efficient and computationally cost-effective fragility assessment framework. The proposed methodology enhances the accuracy and efficiency of fragility models while providing a viable alternative to traditional numerical approaches. Moreover, its applicability extends to diverse structural systems and seismic vulnerability assessments. It offers a valuable tool for earthquake engineering and risk-informed decision-making in seismic-prone regions. However, as with all data-driven models, the framework's performance depends on the quality and diversity of training data, necessitating potential hyperparameter adjustments for structures with significantly different characteristics. Addressing these limitations can provide valuable insights for future research in seismic risk analysis.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"182 ","pages":"Article 113485"},"PeriodicalIF":7.2,"publicationDate":"2025-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144654294","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 single image deraining algorithm guided by text generation based on depth information conditions","authors":"Xing Wei,&nbsp;Xiufen Ye,&nbsp;Xinkui Mei,&nbsp;Junting Wang,&nbsp;Heming Ma","doi":"10.1016/j.asoc.2025.113506","DOIUrl":"10.1016/j.asoc.2025.113506","url":null,"abstract":"<div><div>Currently, image denoising algorithms based on text-to-image diffusion models often encounter issues with disordered internal structure layouts and discrepancies in detail when generating high-resolution images. To address these issues, we proposed a single image deraining algorithm guided by text generation based on depth information conditions. We designed a depth information encoder aimed at leveraging the depth information in rainy images to enhance the spatial mapping between text-to-image and image-to-text, thereby improving the internal structural layout of the generated images. To make the texture details of the generated image domain more similar to those of the original image domain, we designed a Cross Attention module that uses difference information to make the images in both domains more similar, thereby enhancing the guidance of existing deraining algorithms. Experimental results on multiple benchmark datasets demonstrate that the proposed algorithm outperforms state-of-the-art image deraining methods in both visual quality and quantitative performance. On average, it achieves an improvement of 0.46 in SSIM and 0.79 dB in PSNR, effectively removing rain streaks while preserving fine image details and maintaining structural consistency. We will release our code on <span><span>Github</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50737,"journal":{"name":"Applied Soft Computing","volume":"181 ","pages":"Article 113506"},"PeriodicalIF":7.2,"publicationDate":"2025-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144557211","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}