IF 4.5 Q2 COMPUTER SCIENCE, THEORY & METHODS

Array

Pub Date : 2026-01-21 DOI: 10.1016/j.array.2026.100683

Zhiyuan Liang, Pengtao He, Xiaofei Huang, Xiaolei Zhao, Bin Wei, Weili Zeng

During the transportation of cigarette accessories, AGV trolleys often encounter challenges such as complex environments, variable paths, and dynamic obstacles. Traditional path planning methods - such as laser navigation or two-dimensional code navigation - often exhibit limitations including delayed response, poor adaptability, and difficulties in achieving global optimization when dealing with illumination variations, path deviations, and multi-task concurrency. To address these issues, this study investigates a vision-guided intelligent selection algorithm for the feeding path of AGVs in cigarette accessory transportation. The algorithm employs a vision-based AGV path deviation recognition method to identify deviations in the feeding path during cigarette accessory delivery. A multi-step grid approach is utilized to model the feeding environment as a grid map. Within this grid map, an intelligent feeding path selection model based on the pigeon-inspired optimizer (PIO) is applied. An objective function is designed to achieve the shortest path for safely reaching the feeding destination after path deviation. The PIO algorithm optimizes the path to enable vision-guided intelligent selection of the feeding path for AGVs transporting cigarette accessories. Experimental results demonstrate that the proposed algorithm can dynamically plan the optimal feeding path, reduce travel distance, and improve feeding efficiency in vision-guided feeding path selection for AGVs delivering cigarette accessories.

在卷烟配件的运输过程中，AGV小车经常会遇到复杂的环境、多变的路径和动态障碍物等挑战。传统的路径规划方法，如激光导航或二维码导航，在处理光照变化、路径偏差和多任务并发时，往往表现出响应延迟、适应性差、难以实现全局优化等局限性。针对这些问题，研究了一种视觉引导的卷烟配件运输agv投料路径智能选择算法。该算法采用基于视觉的AGV路径偏差识别方法，对卷烟配件投放过程中进料路径的偏差进行识别。采用多步网格方法将饲养环境建模为网格图。在该网格图中，应用了基于鸽子启发优化器（PIO）的智能投食路径选择模型。设计了一个目标函数，以实现路径偏离后安全到达馈送目的地的最短路径。PIO算法对路径进行优化，使运输卷烟配件的agv能够在视觉引导下智能选择投料路径。实验结果表明，该算法可以动态规划最优馈送路径，缩短馈送距离，提高馈送效率。

{"title":"Intelligent selection algorithm of vision guided feeding path for AGV car transporting cigarette accessories","authors":"Zhiyuan Liang, Pengtao He, Xiaofei Huang, Xiaolei Zhao, Bin Wei, Weili Zeng","doi":"10.1016/j.array.2026.100683","DOIUrl":"10.1016/j.array.2026.100683","url":null,"abstract":"<div><div>During the transportation of cigarette accessories, AGV trolleys often encounter challenges such as complex environments, variable paths, and dynamic obstacles. Traditional path planning methods - such as laser navigation or two-dimensional code navigation - often exhibit limitations including delayed response, poor adaptability, and difficulties in achieving global optimization when dealing with illumination variations, path deviations, and multi-task concurrency. To address these issues, this study investigates a vision-guided intelligent selection algorithm for the feeding path of AGVs in cigarette accessory transportation. The algorithm employs a vision-based AGV path deviation recognition method to identify deviations in the feeding path during cigarette accessory delivery. A multi-step grid approach is utilized to model the feeding environment as a grid map. Within this grid map, an intelligent feeding path selection model based on the pigeon-inspired optimizer (PIO) is applied. An objective function is designed to achieve the shortest path for safely reaching the feeding destination after path deviation. The PIO algorithm optimizes the path to enable vision-guided intelligent selection of the feeding path for AGVs transporting cigarette accessories. Experimental results demonstrate that the proposed algorithm can dynamically plan the optimal feeding path, reduce travel distance, and improve feeding efficiency in vision-guided feeding path selection for AGVs delivering cigarette accessories.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"29 ","pages":"Article 100683"},"PeriodicalIF":4.5,"publicationDate":"2026-01-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146073943","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

An automated multi-scale and multi-contextual MobileNetv3 for malware detection based on IoT 基于物联网的自动多尺度和多上下文MobileNetv3恶意软件检测

IF 4.5 Q2 COMPUTER SCIENCE, THEORY & METHODS

Array

Pub Date : 2026-01-20 DOI: 10.1016/j.array.2026.100681

Sidra Javed , Guowei Wu , Hamza Javed , Osama A. Khashan , Haseeb Hassan , Anwar Ghani

Malware detection is a crucial aspect of cybersecurity, aimed at identifying and mitigating malicious software that poses threats to systems and networks. Traditional malware detection methods face challenges in terms of both detection accuracy and computational cost, as deep learning models can be resource-intensive and difficult to deploy in real-time environments. This paper introduces the novel MSMC-MobileNet (Multi-Scale and Multi-Contextual MobileNet) malware detection and classification model, designed to address the challenges of accuracy and computational cost. First, MobileNetv3 is used to extract features from the dataset. To enhance feature extraction, the SE (Squeeze-and-Excitation) module is integrated, focusing on the region of interest using an attention mechanism. The multiscale and multicontextual features are extracted using the ASPP (Atrous Spatial Pyramid Pooling) and FPP (Feature Pyramid Pooling) modules. Channel-wise pruning is applied to the ASPP and FPP modules, reducing computational cost. The model is evaluated on the publicly available Malimg and MaleVis datasets. The proposed MSMC-MobileNet model achieves impressive performance with 92.37% accuracy, 96.54% precision, 95.84% recall, 95.47% F1 score, and 98.59% AUC on the Malimg dataset. On the MaleVis dataset, the model yields 95.08% accuracy, 98.33% precision, 97.9% recall, 98.15% F1 score, and 96.98% AUC. When both datasets are combined, the MSMC-MobileNet achieves 98.79% accuracy, 99.84% precision, 99.73% recall, 99.89% F1 score, and 1.00 AUC. Despite its high accuracy, the model remains computationally efficient, outperforming state-of-the-art methods in both detection performance and computational cost.

恶意软件检测是网络安全的一个重要方面，旨在识别和减轻对系统和网络构成威胁的恶意软件。传统的恶意软件检测方法在检测精度和计算成本方面都面临挑战，因为深度学习模型可能是资源密集型的，并且难以在实时环境中部署。本文介绍了一种新的MSMC-MobileNet （Multi-Scale and Multi-Contextual MobileNet）恶意软件检测与分类模型，旨在解决准确性和计算成本方面的挑战。首先，使用MobileNetv3从数据集中提取特征。为了增强特征提取，集成了SE （Squeeze-and-Excitation）模块，使用注意机制聚焦于感兴趣的区域。利用空间金字塔池（ASPP）和特征金字塔池（FPP）模块提取多尺度和多上下文特征。在ASPP和FPP模块中应用了通道方向剪枝，减少了计算成本。该模型在公开可用的Malimg和MaleVis数据集上进行了评估。本文提出的MSMC-MobileNet模型在Malimg数据集上的准确率为92.37%，精密度为96.54%，召回率为95.84%，F1分数为95.47%，AUC为98.59%。在MaleVis数据集上，该模型的准确率为95.08%，精度为98.33%，召回率为97.9%，F1得分为98.15%，AUC为96.98%。当两个数据集结合在一起时，MSMC-MobileNet的准确率为98.79%，精密度为99.84%，召回率为99.73%，F1分数为99.89%，AUC为1.00。尽管该模型具有很高的精度，但其计算效率仍然很高，在检测性能和计算成本方面都优于最先进的方法。

{"title":"An automated multi-scale and multi-contextual MobileNetv3 for malware detection based on IoT","authors":"Sidra Javed , Guowei Wu , Hamza Javed , Osama A. Khashan , Haseeb Hassan , Anwar Ghani","doi":"10.1016/j.array.2026.100681","DOIUrl":"10.1016/j.array.2026.100681","url":null,"abstract":"<div><div>Malware detection is a crucial aspect of cybersecurity, aimed at identifying and mitigating malicious software that poses threats to systems and networks. Traditional malware detection methods face challenges in terms of both detection accuracy and computational cost, as deep learning models can be resource-intensive and difficult to deploy in real-time environments. This paper introduces the novel MSMC-MobileNet (Multi-Scale and Multi-Contextual MobileNet) malware detection and classification model, designed to address the challenges of accuracy and computational cost. First, MobileNetv3 is used to extract features from the dataset. To enhance feature extraction, the SE (Squeeze-and-Excitation) module is integrated, focusing on the region of interest using an attention mechanism. The multiscale and multicontextual features are extracted using the ASPP (Atrous Spatial Pyramid Pooling) and FPP (Feature Pyramid Pooling) modules. Channel-wise pruning is applied to the ASPP and FPP modules, reducing computational cost. The model is evaluated on the publicly available Malimg and MaleVis datasets. The proposed MSMC-MobileNet model achieves impressive performance with 92.37% accuracy, 96.54% precision, 95.84% recall, 95.47% F1 score, and 98.59% AUC on the Malimg dataset. On the MaleVis dataset, the model yields 95.08% accuracy, 98.33% precision, 97.9% recall, 98.15% F1 score, and 96.98% AUC. When both datasets are combined, the MSMC-MobileNet achieves 98.79% accuracy, 99.84% precision, 99.73% recall, 99.89% F1 score, and 1.00 AUC. Despite its high accuracy, the model remains computationally efficient, outperforming state-of-the-art methods in both detection performance and computational cost.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"29 ","pages":"Article 100681"},"PeriodicalIF":4.5,"publicationDate":"2026-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146034470","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A data-driven comparative analysis of Agile and Waterfall methodologies: Predicting cost and schedule variances using statistical and machine learning approaches 敏捷和瀑布方法的数据驱动比较分析：使用统计和机器学习方法预测成本和进度差异

IF 4.5 Q2 COMPUTER SCIENCE, THEORY & METHODS

Array

Pub Date : 2026-01-19 DOI: 10.1016/j.array.2025.100665

Utkarsh Mishra, Narayanan Ganesh

Project management methodologies like Agile, Waterfall, etc., play an impactful role in key performance indicators of the project, such as cost variance, schedule variance, etc. In this work, we deep dive into these variances with data-driven techniques and discover machine learning models for cost estimation. To demonstrate the efficacy of our approach, we processed a dataset with Agile and Waterfall project attributes which was collected by means of survey conducted online about 100 developers from various companies. We had through categorical encoding, statistical analysis, hypothesis testing, and predictive modeling to predict and compare the projects which can be successful. In the initial stages of Exploratory Data Analysis (EDA), it can be observed that the distribution of cost and schedule variance is not uniform across the waterfall and agile approaches, whereby the mean cost and schedule variances is 2.14 and SD is 1.32 for Agile projects and the mean cost and schedule variances for waterfall projects is higher at 3.87 with SD of 1.89. A T-test conducted to compare the methodologies results in a test statistic of −4.72 and a p-value of 0.00002, indicating a statistically significant difference in cost and schedule variances between Agile and Waterfall projects. Additionally, the use of project attributes to train a linear regression model for predicting cost variance and schedule variance for both waterfall and agile approaches achieves an average MAE of 0.98 and an average MSE of 1.54, indicating moderate predictive accuracy in the models. They emphasize that, on average, Agile projects have a lower cost and schedule variance than Waterfall projects and strengthen the impact of the project methodology on effort deviations. The study highlights the role of predictive analytics in project management and advocates the adoption of machine learning for more accurate cost estimation. The next step is to investigate more advanced modeling techniques and the use of additional project parameters to improve predictive performance and project planning.

项目管理方法，如敏捷、瀑布等，在项目的关键绩效指标（如成本差异、进度差异等）中发挥着重要作用。在这项工作中，我们使用数据驱动技术深入研究这些差异，并发现用于成本估算的机器学习模型。为了证明我们方法的有效性，我们处理了一个包含敏捷和瀑布项目属性的数据集，该数据集是通过对来自不同公司的100名开发人员进行在线调查收集的。我们通过分类编码、统计分析、假设检验和预测建模来预测和比较可能成功的项目。在探索性数据分析（Exploratory Data Analysis， EDA）的初始阶段，可以观察到瀑布式和敏捷式项目的成本和进度方差分布并不均匀，敏捷式项目的平均成本和进度方差为2.14，SD为1.32，而瀑布式项目的平均成本和进度方差更高，为3.87，SD为1.89。进行t检验以比较方法的结果，检验统计量为- 4.72，p值为0.00002，表明在敏捷和瀑布项目之间的成本和进度差异在统计上有显著差异。此外，使用项目属性来训练线性回归模型来预测瀑布方法和敏捷方法的成本方差和进度方差，平均MAE为0.98，平均MSE为1.54，表明模型的预测精度适中。他们强调，平均而言，敏捷项目比瀑布项目具有更低的成本和进度差异，并加强了项目方法对工作偏差的影响。该研究强调了预测分析在项目管理中的作用，并提倡采用机器学习来进行更准确的成本估算。下一步是研究更高级的建模技术和使用额外的项目参数来改进预测性能和项目计划。

{"title":"A data-driven comparative analysis of Agile and Waterfall methodologies: Predicting cost and schedule variances using statistical and machine learning approaches","authors":"Utkarsh Mishra, Narayanan Ganesh","doi":"10.1016/j.array.2025.100665","DOIUrl":"10.1016/j.array.2025.100665","url":null,"abstract":"<div><div>Project management methodologies like Agile, Waterfall, etc., play an impactful role in key performance indicators of the project, such as cost variance, schedule variance, etc. In this work, we deep dive into these variances with data-driven techniques and discover machine learning models for cost estimation. To demonstrate the efficacy of our approach, we processed a dataset with Agile and Waterfall project attributes which was collected by means of survey conducted online about 100 developers from various companies. We had through categorical encoding, statistical analysis, hypothesis testing, and predictive modeling to predict and compare the projects which can be successful. In the initial stages of Exploratory Data Analysis (EDA), it can be observed that the distribution of cost and schedule variance is not uniform across the waterfall and agile approaches, whereby the mean cost and schedule variances is 2.14 and SD is 1.32 for Agile projects and the mean cost and schedule variances for waterfall projects is higher at 3.87 with SD of 1.89. A T-test conducted to compare the methodologies results in a test statistic of −4.72 and a p-value of 0.00002, indicating a statistically significant difference in cost and schedule variances between Agile and Waterfall projects. Additionally, the use of project attributes to train a linear regression model for predicting cost variance and schedule variance for both waterfall and agile approaches achieves an average MAE of 0.98 and an average MSE of 1.54, indicating moderate predictive accuracy in the models. They emphasize that, on average, Agile projects have a lower cost and schedule variance than Waterfall projects and strengthen the impact of the project methodology on effort deviations. The study highlights the role of predictive analytics in project management and advocates the adoption of machine learning for more accurate cost estimation. The next step is to investigate more advanced modeling techniques and the use of additional project parameters to improve predictive performance and project planning.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"29 ","pages":"Article 100665"},"PeriodicalIF":4.5,"publicationDate":"2026-01-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146034469","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Ladderpath: An efficient algorithm for revealing nested hierarchy in sequences Ladderpath：在序列中显示嵌套层次结构的有效算法

IF 4.5 Q2 COMPUTER SCIENCE, THEORY & METHODS

Array

Pub Date : 2026-01-09 DOI: 10.1016/j.array.2025.100663

Jingwen Zhang , Xiao Xie , Xiaodong Deng , Jing Wang , Xiaojun Hu , Yiping Wang , Hu Zhu , Fengyao Zhai , Yu Liu

Ladderpath, rooted in Algorithmic Information Theory (AIT), uncovers nested and hierarchical structures in symbolic sequences through minimal compositional reconstruction. It approximates Kolmogorov complexity by identifying reusable subsequences that enable efficient reconstruction of complex sequences. The proposed algorithm improves upon earlier implementations by introducing key optimizations in substring enumeration and reuse filtering, allowing it to scale to sequence systems with tens or even hundreds of millions of characters. Ladderpath produces a standardized JSON format that encodes compositional dependencies and hierarchies, and supports a variety of downstream tasks, including compression, shared motif extraction, cross-sequence similarity analysis, and structural visualization. Its domain-agnostic design enables broad applicability across areas such as genomics, natural language, symbolic computation, and program analysis. Beyond providing a practical approximation of complexity, Ladderpath also offers structural insight into the modular grammar of sequences, pointing to a deeper connection between algorithmic complexity and compositional hierarchies observed in real-world data.

Ladderpath根植于算法信息理论（AIT），通过最小的组成重构揭示符号序列中的嵌套和分层结构。它通过识别可重用的子序列来近似Kolmogorov复杂度，从而能够有效地重建复杂序列。提出的算法通过在子字符串枚举和重用过滤中引入关键优化，改进了早期的实现，允许它扩展到具有数千万甚至数亿个字符的序列系统。Ladderpath生成一种标准化的JSON格式，对组合依赖关系和层次结构进行编码，并支持各种下游任务，包括压缩、共享基序提取、交叉序列相似性分析和结构可视化。它的领域不可知的设计使其在基因组学、自然语言、符号计算和程序分析等领域具有广泛的适用性。除了提供实际的复杂性近似值之外，Ladderpath还提供了对序列模块化语法的结构洞察，指出了在现实世界数据中观察到的算法复杂性和组合层次结构之间的更深层次的联系。

{"title":"Ladderpath: An efficient algorithm for revealing nested hierarchy in sequences","authors":"Jingwen Zhang , Xiao Xie , Xiaodong Deng , Jing Wang , Xiaojun Hu , Yiping Wang , Hu Zhu , Fengyao Zhai , Yu Liu","doi":"10.1016/j.array.2025.100663","DOIUrl":"10.1016/j.array.2025.100663","url":null,"abstract":"<div><div>Ladderpath, rooted in Algorithmic Information Theory (AIT), uncovers nested and hierarchical structures in symbolic sequences through minimal compositional reconstruction. It approximates Kolmogorov complexity by identifying reusable subsequences that enable efficient reconstruction of complex sequences. The proposed algorithm improves upon earlier implementations by introducing key optimizations in substring enumeration and reuse filtering, allowing it to scale to sequence systems with tens or even hundreds of millions of characters. Ladderpath produces a standardized JSON format that encodes compositional dependencies and hierarchies, and supports a variety of downstream tasks, including compression, shared motif extraction, cross-sequence similarity analysis, and structural visualization. Its domain-agnostic design enables broad applicability across areas such as genomics, natural language, symbolic computation, and program analysis. Beyond providing a practical approximation of complexity, Ladderpath also offers structural insight into the modular grammar of sequences, pointing to a deeper connection between algorithmic complexity and compositional hierarchies observed in real-world data.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"29 ","pages":"Article 100663"},"PeriodicalIF":4.5,"publicationDate":"2026-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145973362","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Design and implementation of a scalable microservice-based cloud computing service for stochastic fluid flow modeling 设计和实现一个可扩展的基于微服务的云计算服务，用于随机流体流动建模

IF 4.5 Q2 COMPUTER SCIENCE, THEORY & METHODS

Array

Pub Date : 2026-01-09 DOI: 10.1016/j.array.2026.100679

Zhanars Abdiramanov

This paper presents the design and performance evaluation of a scalable microservice-based cloud computing framework for distributed stochastic fluid flow simulations. The system integrates Docker containerization, NGINX load balancing, and Apache Kafka for asynchronous task coordination and efficient workload distribution across computational nodes. The backend combines PHP and Julia for orchestration and computation, with PostgreSQL managing task and result data. Benchmark experiments demonstrate near-linear scalability and stable performance under varying loads, confirming the system’s suitability for high-performance scientific computing. The proposed framework advances parallel and distributed computing by introducing a modular, reproducible architecture for executing complex stochastic models in cloud environments.

本文提出了一个可扩展的基于微服务的分布式随机流体流动模拟云计算框架的设计和性能评估。该系统集成了Docker容器化、NGINX负载均衡和Apache Kafka，用于异步任务协调和跨计算节点的高效工作负载分配。后端结合PHP和Julia进行编排和计算，PostgreSQL管理任务和结果数据。基准实验证明了该系统在不同负载下的近线性可扩展性和稳定性能，验证了该系统适用于高性能科学计算。该框架通过引入模块化、可重复的架构，在云环境中执行复杂的随机模型，从而推进了并行和分布式计算。

引用次数: 0

Robust multi-scale LSTM for reliable and intelligent photovoltaic power forecasting 基于鲁棒多尺度LSTM的可靠智能光伏发电预测

IF 4.5 Q2 COMPUTER SCIENCE, THEORY & METHODS

Array

Pub Date : 2026-01-08 DOI: 10.1016/j.array.2026.100678

Chongchong Xu , Bin Zhao , Yujie Zhao

Photovoltaic (PV) power generation, which playing a crucial role in promoting sustainable development and mitigating climate change, has become a key catalyst in the worldwide energy shift. However, its extensive integration into power grids presents issues for stability and efficient utilization, as PV output is highly dependent on meteorological conditions. The inherent variability and randomness of PV power time series often degrade forecasting performance. To address these challenges, a prediction approach based on a Robust Multi-scale Long Short-Term Memory Network (RMS-LSTM) is proposed in this study. We aim to construct a model which is able to capture both short-term and long-term trends. Specifically, we also employs correlation analysis and Robust Principal Component Analysis (RPCA) for feature selection and dimensionality reduction. On this basis, the refined training data inputs the multi-scale LSTM framework, aiming to derive the potential features. Then, we can employ the trained model to attain high-precision predictions of power output. The PV power data from Guoneng serves as a validation dataset to evaluate the performance of RMS-LSTM. It is shown that the Root Mean Square Error (RMSE) is 4.9482 and Mean Absolute Error (MAE) is 2.5246. Compared with benchmark methods, the proposed framework has advantages in prediction accuracy and robustness, verifying its effectiveness in PV power forecasting.

光伏发电在促进可持续发展和减缓气候变化方面发挥着至关重要的作用，已成为全球能源转型的关键催化剂。然而，由于光伏发电产量高度依赖气象条件，因此其与电网的广泛整合带来了稳定性和有效利用方面的问题。光伏发电时间序列固有的可变性和随机性往往会降低预测效果。为了解决这些问题，本研究提出了一种基于鲁棒多尺度长短期记忆网络（RMS-LSTM）的预测方法。我们的目标是构建一个能够捕捉短期和长期趋势的模型。具体而言，我们还使用相关分析和鲁棒主成分分析（RPCA）进行特征选择和降维。在此基础上，将精炼后的训练数据输入到多尺度LSTM框架中，旨在推导出潜在特征。然后，我们可以使用训练好的模型来获得高精度的功率输出预测。国能光伏发电数据作为验证数据，用于评估RMS-LSTM的性能。结果表明，均方根误差（RMSE）为4.9482，平均绝对误差（MAE）为2.5246。与基准方法相比，该框架在预测精度和鲁棒性方面具有优势，验证了其在光伏发电功率预测中的有效性。

{"title":"Robust multi-scale LSTM for reliable and intelligent photovoltaic power forecasting","authors":"Chongchong Xu , Bin Zhao , Yujie Zhao","doi":"10.1016/j.array.2026.100678","DOIUrl":"10.1016/j.array.2026.100678","url":null,"abstract":"<div><div>Photovoltaic (PV) power generation, which playing a crucial role in promoting sustainable development and mitigating climate change, has become a key catalyst in the worldwide energy shift. However, its extensive integration into power grids presents issues for stability and efficient utilization, as PV output is highly dependent on meteorological conditions. The inherent variability and randomness of PV power time series often degrade forecasting performance. To address these challenges, a prediction approach based on a Robust Multi-scale Long Short-Term Memory Network (RMS-LSTM) is proposed in this study. We aim to construct a model which is able to capture both short-term and long-term trends. Specifically, we also employs correlation analysis and Robust Principal Component Analysis (RPCA) for feature selection and dimensionality reduction. On this basis, the refined training data inputs the multi-scale LSTM framework, aiming to derive the potential features. Then, we can employ the trained model to attain high-precision predictions of power output. The PV power data from Guoneng serves as a validation dataset to evaluate the performance of RMS-LSTM. It is shown that the Root Mean Square Error (RMSE) is 4.9482 and Mean Absolute Error (MAE) is 2.5246. Compared with benchmark methods, the proposed framework has advantages in prediction accuracy and robustness, verifying its effectiveness in PV power forecasting.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"29 ","pages":"Article 100678"},"PeriodicalIF":4.5,"publicationDate":"2026-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145973364","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Chosen-plaintext attacks on image ciphers based on nonlinear dynamical systems 基于非线性动态系统的图像密码的选择明文攻击

IF 4.5 Q2 COMPUTER SCIENCE, THEORY & METHODS

Array

Pub Date : 2026-01-08 DOI: 10.1016/j.array.2025.100642

Mohammad Mazyad Hazzazi , Dawood Shah , Mansoor Alghamdi , Ibrahim S. Alkhazi , Naveed Riaz

Recent advances in the fields of wireless communication networks and digital technologies have resulted in a significant increase in multimedia data sharing, especially digital images. Digital images often carry sensitive and personal information, so ensuring their security has become a critical concern. To address these challenges, the cryptography community has developed various encryption techniques, many of which rely on nonlinear dynamical systems and employ processes such as pixel permutation and diffusion to secure images. In this article, we have analyzed various security vulnerabilities in image encryption schemes available in literature that are based on pixel permutation and substitution. The study is divided into two main parts. First, we identify four types of image encryption schemes based on nonlinear dynamical systems and demonstrate their susceptibility to chosen-plaintext attacks. Secondly, we implement a substitution-permutation based image encryption scheme and show that it can be distinguished using a small number of chosen plaintext pairs in a distinguishability game. Furthermore, we prove that an attacker can recover the original plaintext by leveraging knowledge of a few chosen plaintext-ciphertext pairs, despite that the implemented scheme successfully passes all security evaluation tests proposed in the literature for assessing the cryptographic strength of image encryption schemes.

无线通信网络和数字技术领域的最新进展导致多媒体数据共享，特别是数字图像的显著增加。数字图像通常携带敏感和个人信息，因此确保其安全已成为一个关键问题。为了应对这些挑战，密码学社区开发了各种加密技术，其中许多技术依赖于非线性动态系统，并采用诸如像素排列和扩散等过程来保护图像。在本文中，我们分析了文献中基于像素置换和替换的图像加密方案中的各种安全漏洞。本研究分为两个主要部分。首先，我们识别了四种基于非线性动态系统的图像加密方案，并证明了它们对选择明文攻击的敏感性。其次，我们实现了一种基于替换置换的图像加密方案，并证明了它可以在可区分博弈中使用少量选择的明文对进行区分。此外，我们证明了攻击者可以利用所选的几个明文-密文对的知识来恢复原始明文，尽管实现的方案成功地通过了文献中为评估图像加密方案的加密强度而提出的所有安全评估测试。

{"title":"Chosen-plaintext attacks on image ciphers based on nonlinear dynamical systems","authors":"Mohammad Mazyad Hazzazi , Dawood Shah , Mansoor Alghamdi , Ibrahim S. Alkhazi , Naveed Riaz","doi":"10.1016/j.array.2025.100642","DOIUrl":"10.1016/j.array.2025.100642","url":null,"abstract":"<div><div>Recent advances in the fields of wireless communication networks and digital technologies have resulted in a significant increase in multimedia data sharing, especially digital images. Digital images often carry sensitive and personal information, so ensuring their security has become a critical concern. To address these challenges, the cryptography community has developed various encryption techniques, many of which rely on nonlinear dynamical systems and employ processes such as pixel permutation and diffusion to secure images. In this article, we have analyzed various security vulnerabilities in image encryption schemes available in literature that are based on pixel permutation and substitution. The study is divided into two main parts. First, we identify four types of image encryption schemes based on nonlinear dynamical systems and demonstrate their susceptibility to chosen-plaintext attacks. Secondly, we implement a substitution-permutation based image encryption scheme and show that it can be distinguished using a small number of chosen plaintext pairs in a distinguishability game. Furthermore, we prove that an attacker can recover the original plaintext by leveraging knowledge of a few chosen plaintext-ciphertext pairs, despite that the implemented scheme successfully passes all security evaluation tests proposed in the literature for assessing the cryptographic strength of image encryption schemes.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"29 ","pages":"Article 100642"},"PeriodicalIF":4.5,"publicationDate":"2026-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146034354","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Improve semantic similarity based on statistical approach and LLM based transformer model for extractive summarization 基于统计方法和基于LLM的抽取摘要变压器模型提高语义相似度

IF 4.5 Q2 COMPUTER SCIENCE, THEORY & METHODS

Array

Pub Date : 2026-01-07 DOI: 10.1016/j.array.2025.100671

Sutriawan , Supriadi Rustad , Guruh Fajar Shidik , Pujiono

This study addresses the need for effective extractive text summarization methods specifically designed for Indonesian news texts, driven by the increasing volume of digital information and persistent challenges related to semantic drift and coherence instability. A key challenge in extractive summarization lies in maintaining semantic fidelity while preserving informative content and ensuring computational efficiency, particularly for morphologically complex languages such as Indonesian. To address this issue, this research proposes a hybrid extractive summarization approach that integrates statistical methods, namely Term Frequency–Inverse Document Frequency (TF-IDF) and Latent Semantic Analysis (LSA), with Indonesian-specific transformer models, including IndoBERT and GPT, to enhance semantic similarity measurement during sentence selection. The proposed framework employs a weighted hybrid formulation (α·TF-IDF + β·LSA + γ·IndoBERT + δ·GPT), combined with cosine similarity computation and diversity-aware sentence selection. The approach is evaluated under a consistent experimental setting using the Indonesian subset of the XL-Sum dataset, with experiments conducted on TPU V2-8 infrastructure. Experimental results demonstrate competitive performance across the evaluated baseline components. The TF-IDF + LSA model achieved the highest ROUGE-1 score of 63.76 %, while the TF-IDF + Cosine Similarity model showed balanced performance across ROUGE metrics with a cosine similarity score of 0.8538. IndoBERT + Cosine Similarity and GPT + Cosine Similarity achieved higher semantic similarity scores, with GPT reaching 0.8677. Overall, the proposed approach demonstrates substantial improvements over Indonesian extractive summarization baselines under the same dataset and evaluation protocol, highlighting the effectiveness of hybrid statistical-transformer integration for low-resource language summarization.

由于数字信息量的增加以及语义漂移和连贯不稳定性带来的持续挑战，本研究解决了为印度尼西亚新闻文本设计的有效提取文本摘要方法的需求。提取摘要的一个关键挑战在于保持语义保真度，同时保留信息内容并确保计算效率，特别是对于形态学复杂的语言，如印尼语。为了解决这一问题，本研究提出了一种混合提取摘要方法，该方法将统计方法，即术语频率-逆文档频率（TF-IDF）和潜在语义分析（LSA）与印度尼西亚特定的转换模型（包括IndoBERT和GPT）相结合，以增强句子选择过程中的语义相似度测量。该框架采用加权混合公式（α·TF-IDF + β·LSA + γ·IndoBERT + δ·GPT），结合余弦相似度计算和多样性感知句子选择。使用xml - sum数据集的印度尼西亚子集在一致的实验设置下对该方法进行了评估，并在TPU V2-8基础设施上进行了实验。实验结果表明，在评估的基线组件中具有竞争力。TF-IDF + LSA模型的ROUGE-1得分最高，为63.76%，而TF-IDF +余弦相似度模型在ROUGE指标上表现均衡，余弦相似度得分为0.8538。IndoBERT +余弦相似度和GPT +余弦相似度的语义相似度得分更高，GPT达到0.8677。总的来说，在相同的数据集和评估协议下，所提出的方法比印度尼西亚提取摘要基线有了实质性的改进，突出了混合统计-转换集成用于低资源语言摘要的有效性。

{"title":"Improve semantic similarity based on statistical approach and LLM based transformer model for extractive summarization","authors":"Sutriawan , Supriadi Rustad , Guruh Fajar Shidik , Pujiono","doi":"10.1016/j.array.2025.100671","DOIUrl":"10.1016/j.array.2025.100671","url":null,"abstract":"<div><div>This study addresses the need for effective extractive text summarization methods specifically designed for Indonesian news texts, driven by the increasing volume of digital information and persistent challenges related to semantic drift and coherence instability. A key challenge in extractive summarization lies in maintaining semantic fidelity while preserving informative content and ensuring computational efficiency, particularly for morphologically complex languages such as Indonesian. To address this issue, this research proposes a hybrid extractive summarization approach that integrates statistical methods, namely Term Frequency–Inverse Document Frequency (TF-IDF) and Latent Semantic Analysis (LSA), with Indonesian-specific transformer models, including IndoBERT and GPT, to enhance semantic similarity measurement during sentence selection. The proposed framework employs a weighted hybrid formulation (α·TF-IDF + β·LSA + γ·IndoBERT + δ·GPT), combined with cosine similarity computation and diversity-aware sentence selection. The approach is evaluated under a consistent experimental setting using the Indonesian subset of the XL-Sum dataset, with experiments conducted on TPU V2-8 infrastructure. Experimental results demonstrate competitive performance across the evaluated baseline components. The TF-IDF + LSA model achieved the highest ROUGE-1 score of 63.76 %, while the TF-IDF + Cosine Similarity model showed balanced performance across ROUGE metrics with a cosine similarity score of 0.8538. IndoBERT + Cosine Similarity and GPT + Cosine Similarity achieved higher semantic similarity scores, with GPT reaching 0.8677. Overall, the proposed approach demonstrates substantial improvements over Indonesian extractive summarization baselines under the same dataset and evaluation protocol, highlighting the effectiveness of hybrid statistical-transformer integration for low-resource language summarization.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"29 ","pages":"Article 100671"},"PeriodicalIF":4.5,"publicationDate":"2026-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145921317","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

GPAT: Enhancing point cloud registration with local-global Integration and Geo-positional awareness GPAT：利用局部-全局集成和地理位置感知增强点云配准

IF 4.5 Q2 COMPUTER SCIENCE, THEORY & METHODS

Array

Pub Date : 2026-01-07 DOI: 10.1016/j.array.2025.100660

Zijian Wang, Zhenbao Wang, Xin Liu, Xinxi Xu, Chen Su, Xiuguo Zhao

Point cloud registration, essential for applications like autonomous driving and 3D reconstruction, remains challenging in scenarios with low overlap and incomplete data. Although learning-based methods have shown promise, existing keypoint-based and coarse-to-fine approaches can struggle with ambiguous matching due to similar local patches in non-overlapping regions, particularly under low overlap conditions. This study introduces GPAT(Geometric-Positional Attention and Local-Global Transformer Networks), a novel hierarchical deep learning framework designed to address these limitations by effectively integrating local geometric and positional information with global structural awareness. By leveraging a novel positional encoding and attention mechanisms, GPAT captures the intricate interplay between geometric structures and spatial configurations, enriching the distinctiveness of local features. Simultaneously, it incorporates global geometric consistency to establish robust correspondences, effectively bridging the gap between local and global contexts. The framework further refines these correspondences into dense point matches, ensuring precise registration even under challenging conditions. Extensive experiments on benchmark datasets, including 3DMatch, ModelNet, and KITTI, demonstrate GPAT's competitive performance, especially in low-overlap situations, demonstrating its capability to extract discriminative features and achieve accurate transformations with high robustness.

点云配准对于自动驾驶和3D重建等应用至关重要，但在低重叠和数据不完整的情况下仍然具有挑战性。尽管基于学习的方法已经显示出前景，但现有的基于关键点和粗到精的方法可能会由于在非重叠区域中相似的局部补丁而难以进行模糊匹配，特别是在低重叠条件下。本研究引入了GPAT（几何位置注意和局部-全局变压器网络），这是一种新的分层深度学习框架，旨在通过有效地将局部几何和位置信息与全局结构意识相结合来解决这些限制。通过利用新颖的位置编码和注意机制，GPAT捕获了几何结构和空间配置之间复杂的相互作用，丰富了局部特征的独特性。同时，它结合了全局几何一致性来建立健壮的对应关系，有效地弥合了局部和全局上下文之间的差距。该框架进一步将这些对应细化为密集的点匹配，即使在具有挑战性的条件下也能确保精确的配准。在包括3DMatch、ModelNet和KITTI在内的基准数据集上进行的大量实验证明了GPAT的竞争性能，特别是在低重叠情况下，证明了它能够提取判别特征并实现高鲁棒性的精确转换。

{"title":"GPAT: Enhancing point cloud registration with local-global Integration and Geo-positional awareness","authors":"Zijian Wang, Zhenbao Wang, Xin Liu, Xinxi Xu, Chen Su, Xiuguo Zhao","doi":"10.1016/j.array.2025.100660","DOIUrl":"10.1016/j.array.2025.100660","url":null,"abstract":"<div><div>Point cloud registration, essential for applications like autonomous driving and 3D reconstruction, remains challenging in scenarios with low overlap and incomplete data. Although learning-based methods have shown promise, existing keypoint-based and coarse-to-fine approaches can struggle with ambiguous matching due to similar local patches in non-overlapping regions, particularly under low overlap conditions. This study introduces GPAT(Geometric-Positional Attention and Local-Global Transformer Networks), a novel hierarchical deep learning framework designed to address these limitations by effectively integrating local geometric and positional information with global structural awareness. By leveraging a novel positional encoding and attention mechanisms, GPAT captures the intricate interplay between geometric structures and spatial configurations, enriching the distinctiveness of local features. Simultaneously, it incorporates global geometric consistency to establish robust correspondences, effectively bridging the gap between local and global contexts. The framework further refines these correspondences into dense point matches, ensuring precise registration even under challenging conditions. Extensive experiments on benchmark datasets, including 3DMatch, ModelNet, and KITTI, demonstrate GPAT's competitive performance, especially in low-overlap situations, demonstrating its capability to extract discriminative features and achieve accurate transformations with high robustness.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"29 ","pages":"Article 100660"},"PeriodicalIF":4.5,"publicationDate":"2026-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146034355","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Integrating feature fusion with hybrid optimization for multiple sclerosis MRI classification 融合特征融合与混合优化的多发性硬化MRI分类

IF 4.5 Q2 COMPUTER SCIENCE, THEORY & METHODS

Array

Pub Date : 2026-01-06 DOI: 10.1016/j.array.2025.100677

Nandini Anam , Sharief Basha S , Chiranji Lal Chowdhary

Detecting Multiple Sclerosis (MS) has previously been difficult to identify with MRI scans due to the subtlety and dispersion of lesions, as well as other imaging anomalies. The researchers in this study created a unique hybrid framework that combines two state-of-the-art convolutional neural networks, ResNet-50 and EfficientNet-B7. Additionally, a new hybrid bio-inspired optimization strategy combining the Grey Wolf Optimizer (GWO) and the Genetic Algorithm (GA) is described. The technique simplifies the computations and ensures that the best characteristics are picked. We extracted deep features from both CNNs, used Principal Component Analysis (PCA) to reduce them to high dimensions, and then employed the GA-GWO method to identify suitable features. The enhanced artificial neural network (ANN) classifier outperformed standalone CNN-based models, achieving a maximum accuracy of 90.67 %(refer to table 4). The suggested framework for dependable and comprehensible precision and processing efficacy. This work has the potential to motivate future studies in related areas.

由于病变的微妙性和弥散性以及其他成像异常，以前很难通过MRI扫描来识别多发性硬化症（MS）。在这项研究中，研究人员创建了一个独特的混合框架，结合了两个最先进的卷积神经网络ResNet-50和EfficientNet-B7。此外，还提出了一种将灰狼优化器（GWO）与遗传算法（GA）相结合的混合生物优化策略。该技术简化了计算，并确保了最佳特征的选择。我们从两个cnn中提取深度特征，使用主成分分析（PCA）将其降维到高维，然后使用GA-GWO方法识别合适的特征。增强的人工神经网络（ANN）分类器优于独立的基于cnn的模型，最高准确率达到90.67%（见表4）。建议的框架是可靠和可理解的精度和加工效率。这项工作有可能激发未来相关领域的研究。

引用次数: 0

Array最新文献