IF 4.5

Array Pub Date : 2025-08-21 DOI: 10.1016/j.array.2025.100494

Mohammed Hamzah Alsalihi , Dávid Sztahó

引用次数: 0

MalGEA: A malware analysis framework via matrix factorization based node embedding and graph external attention MalGEA：一个基于矩阵分解的节点嵌入和图外部关注的恶意软件分析框架

IF 4.5

Array Pub Date : 2025-08-19 DOI: 10.1016/j.array.2025.100493

Ruisheng Li , Qilong Zhang , Huimin Shen

{"title":"MalGEA: A malware analysis framework via matrix factorization based node embedding and graph external attention","authors":"Ruisheng Li , Qilong Zhang , Huimin Shen","doi":"10.1016/j.array.2025.100493","DOIUrl":"10.1016/j.array.2025.100493","url":null,"abstract":"<div><div>As one of the major threats in cybersecurity, malware has been growing continuously and steadily. In recent years, researchers have proposed a number of graph representation learning based malware detection methods by leveraging the intrinsic topological features of malware, which has led to considerable development in this area. However, these existing malware studies still have two major limitations. (1) The complex topological structures of malware graphs often result in high computational overhead during feature extraction and processing. (2) Most existing approaches rely on conventional graph neural networks that are not specifically designed for malware classification tasks, leading to suboptimal performance, especially when dealing with minority class samples. To address these problems, we propose MalGEA, a novel malware detection and classification framework based on matrix factorization and graph external attention mechanisms. First, MalGEA extracts function call information from malware and constructs corresponding function call graphs. These graphs are then processed using sparse matrix factorization and spectral propagation to efficiently generate node embeddings. Finally, we employ an graph external attention network to model inter-graph relationships and perform malware detection and classification. To evaluate our approach, we utilized a benchmark malware dataset which contains 6 categories and 35 families, including 50k benign and 50k malicious samples. Experimental results demonstrate that our method significantly outperforms existing node embedding approaches in terms of computational efficiency, while also achieving high accuracy in malware detection and family classification tasks.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"27 ","pages":"Article 100493"},"PeriodicalIF":4.5,"publicationDate":"2025-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144888649","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

Binary2vec:Cross-architecture binary embeddings with global attention-enhanced graph neural networks Binary2vec：基于全局注意力增强图神经网络的跨架构二进制嵌入

IF 4.5

Array Pub Date : 2025-08-18 DOI: 10.1016/j.array.2025.100491

Zhenyu Gao , Lei Xiao , Wei Weng , Qizhen Xu , Baishun Zhou , Longquan Luo

{"title":"Binary2vec:Cross-architecture binary embeddings with global attention-enhanced graph neural networks","authors":"Zhenyu Gao , Lei Xiao , Wei Weng , Qizhen Xu , Baishun Zhou , Longquan Luo","doi":"10.1016/j.array.2025.100491","DOIUrl":"10.1016/j.array.2025.100491","url":null,"abstract":"<div><div>Binary analysis is crucial in the software security domain, supporting tasks such as software plagiarism detection and reverse engineering. However, existing methods either struggle to generalize across hardware architectures or fail to fully capture high-level program semantics. Moreover, in binary similarity analysis, some approaches yield both high precision and high mean squared error, indicating that they assign high similarity scores to both similar and dissimilar binaries. To address these challenges, we propose Binary2vec, a novel framework for constructing cross-architecture binary embeddings. First, Binary2vec leverages the LLVM intermediate representation to achieve cross-architecture compatibility. Then, Binary2vec captures program semantics through a novel graph representation, A-PROGRAML. Finally, A-PROGRAML graph is fed into a graph neural network called GPS with a global attention mechanism to obtain the binary embeddings. To demonstrate its effectiveness, we evaluate Binary2vec on three binary analysis tasks: heterogeneous compute device mapping, optimal thread coarsening factor prediction, and similarity analysis. In heterogeneous compute device mapping and optimal thread coarsening factor prediction, Binary2vec demonstrates better performance than NCC and IR2VEC on average. In similarity analysis, Binary2vec outperforms BinaryAI (the state-of-the-art method) proposed by Tencent Security Keen Lab in cross-architecture scenarios, and works well even with binaries with a small number of functions, where BinaryAI fails.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"27 ","pages":"Article 100491"},"PeriodicalIF":4.5,"publicationDate":"2025-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144864708","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

Double Layered Blockchain-based trust model for secure interest and data forwarding in Vehicular Information Centric Network 基于双层区块链的车辆信息中心网络安全利益和数据转发信任模型

IF 4.5

Array Pub Date : 2025-08-16 DOI: 10.1016/j.array.2025.100490

Sanjeev Kumar Mekala, Satish Anamalamudi, Anil Carie, Murali Krishna Enduri

{"title":"Double Layered Blockchain-based trust model for secure interest and data forwarding in Vehicular Information Centric Network","authors":"Sanjeev Kumar Mekala, Satish Anamalamudi, Anil Carie, Murali Krishna Enduri","doi":"10.1016/j.array.2025.100490","DOIUrl":"10.1016/j.array.2025.100490","url":null,"abstract":"<div><div>Vehicular Information Centric Networks(V-ICNs) which is an alternate to the traditional Vehicular Adhoc Networks (VANETs) is proposed to enable the content-based addressing instead of IP based data access to improve the efficiency of the Vehicular Network. V-ICN is more susceptible to security attacks from several sources because of its wireless, heterogeneous connection style and highly dynamic architecture. In contrast to entity-based security authentication, it is crucial to think about how to safeguard the data’s security. Although reputation based quantification has been used in state-of-the-art research to assess the dependability of interactive data, there are still some problems with the design of safe reputation management systems. This result in low efficiency, inadequate security and unreliable administration. The Double Layered Blockchain (DLB) technique for communication security in vehicle Information-Centric Networks will be presented in this study. It will take into account of both the worldwide reputation of vehicle chain and the one-day local information chain. In this, each vehicle’s activities that are documented in the Local One-day Message Blockchain (LOMB) will be used to update the reputation score of the vehicles on a regular basis. In addition, the proposed work would also include a Secured Neighbourhood Recognition Protocol (SNRP) to introduce flexible connectivity among the vehicle nodes and the blockchain network. Based on the Experimental Analysis, the proposed methodology is outperformed in comparison with the existing models in terms of network throughput, interest success rate, varying interest generation rate and the interest success rate.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"27 ","pages":"Article 100490"},"PeriodicalIF":4.5,"publicationDate":"2025-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144893503","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

NLP based text classification using TF-IDF enabled fine-tuned long short-term memory: An empirical analysis 基于NLP的文本分类使用TF-IDF启用微调长短期记忆：实证分析

IF 4.5

Array Pub Date : 2025-08-14 DOI: 10.1016/j.array.2025.100467

Pratiyush Guleria , Jaroslav Frnda , Parvathaneni Naga Srinivasu

{"title":"NLP based text classification using TF-IDF enabled fine-tuned long short-term memory: An empirical analysis","authors":"Pratiyush Guleria , Jaroslav Frnda , Parvathaneni Naga Srinivasu","doi":"10.1016/j.array.2025.100467","DOIUrl":"10.1016/j.array.2025.100467","url":null,"abstract":"<div><div>The rapid proliferation of information through digital transformation and the widespread use of social networking platforms has significantly increased the speed of information dissemination across urban and rural areas alike. While these platforms have become vital channels for sharing news, advertisements, and crucial updates, they also pose challenges in verifying the authenticity of the information in real-time. Addressing this issue, this study proposes a novel Convolutional Neural Networks (CNN)-Long Short-Term Memory (LSTM) model designed for the classification of fake news articles. A comprehensive dataset covering diverse categories, including government news, Middle East news, US news, left-wing news, and political content, was utilized in this research. Following preprocessing, features were extracted using the Term Frequency-Inverse Document Frequency (TF-IDF) technique, and word embeddings were generated for enhanced semantic representation. The combined CNN-LSTM model leverages the strengths of both architectures, capturing local patterns and long-range dependencies within the data. The experimental results demonstrate that the Fine-Tuned CNN-LSTM model outperforms all precedent approaches across various categories. Notably, the Fine-Tuned CNN-LSTM model achieves the highest accuracy (AC), ranging from 0.57 to 0.68, highlighting its superior classification performance to other precedent approaches, indicating their inefficacy in handling multiple categories.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"27 ","pages":"Article 100467"},"PeriodicalIF":4.5,"publicationDate":"2025-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144860368","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

Agile Development for Safety Assurance of Machine Learning in Autonomous Systems (AgileAMLAS) 自主系统中机器学习安全保障的敏捷开发（AgileAMLAS）

IF 4.5

Array Pub Date : 2025-08-14 DOI: 10.1016/j.array.2025.100482

Victoria J. Hodge, Matt Osborne

{"title":"Agile Development for Safety Assurance of Machine Learning in Autonomous Systems (AgileAMLAS)","authors":"Victoria J. Hodge, Matt Osborne","doi":"10.1016/j.array.2025.100482","DOIUrl":"10.1016/j.array.2025.100482","url":null,"abstract":"<div><div>Recent advances in ML have enabled the development of autonomous cyber–physical systems for a broad range of applications. Using ML, these autonomous systems are able to learn, adapt, and operate with no human intervention. However, this autonomous operation poses a problem when proving that they are acceptably safe. Designers and engineers have traditionally used ‘Waterfall’ or V-model development lifecycles to develop safe systems, but ML engineering requires iteration and adaptation. Iterative development necessitates enhanced lifecycles, augmented methodologies, and the need to systematically integrate rigorous safety assurance with ML development and operation activities. In this paper, we introduce a novel lifecycle, and comprehensive methodology for safely developing, operating, and assuring autonomous systems which use ML. The lifecycle combines Agile software engineering, ML engineering, and a safety engineering framework using iterative and incremental development. This paper provides systematic step-by-step guidelines for developing and deploying ML for autonomous systems using DevOps and MLOps, and for generating compelling safety cases. We have developed and refined our methodology on a recent set of projects undertaken to develop autonomous robots across a variety of domains.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"27 ","pages":"Article 100482"},"PeriodicalIF":4.5,"publicationDate":"2025-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144864709","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 efficient metaheuristic optimization algorithm for optimal power extraction from PV systems under various weather and load-changing conditions 一种有效的元启发式优化算法，用于各种天气和负荷变化条件下光伏系统的最优取电

IF 4.5

Array Pub Date : 2025-08-13 DOI: 10.1016/j.array.2025.100492

Md.Al Imran Fahim, Md.Salah Uddin Yusuf, Monira Islam, Munshi Jawad Ibne Azad

{"title":"An efficient metaheuristic optimization algorithm for optimal power extraction from PV systems under various weather and load-changing conditions","authors":"Md.Al Imran Fahim, Md.Salah Uddin Yusuf, Monira Islam, Munshi Jawad Ibne Azad","doi":"10.1016/j.array.2025.100492","DOIUrl":"10.1016/j.array.2025.100492","url":null,"abstract":"<div><div>Currently, the focus has been shifted towards exploring solar energy due to its environmentally friendly and economic nature. However, the efficiency of photovoltaic (PV) systems can be impacted by factors such as ineffective Global Maxima (GM) tracking, slow response time in tracking, becoming stuck in local maxima, and fluctuations around GM. To address these challenges, a new algorithm called horse herd optimization (HHO) has been applied to the maximum power point tracking (MPPT) controller. The proposed approach has four key features: high efficiency, cheap computing power, rapid MPPT, and zero oscillation. A comprehensive study compares the HHO technique with established methods such as perturb and observe (P&O), modified P&O (MP&O), incremental conductance (IC), Spline MPPT, particle swarm optimization (PSO), grasshopper optimization (GHO), and grey wolf optimization (GWO) across fast-changing irradiance, partial shading, complex partial shading, and load-changing conditions. All models and scenarios were implemented and tested in the MATLAB/Simulink environment. An adaptive search mechanism is integrated into HHO to improve its resilience. The results demonstrate that HHO shows robustness with the highest average tracking efficiency reaching 99.98 % with the least tracking time up to 160 msec while keeping the steady-state oscillation below 0.5 W. According to quantitative, comparative, and statistical results, the HHO-based MPPT performs better by achieving at least 21 % faster tracking time and 16 % faster settling time, and up to 4.4 % increase in power efficiency, which shows the effectiveness of the proposed technique.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"27 ","pages":"Article 100492"},"PeriodicalIF":4.5,"publicationDate":"2025-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144858291","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

IsharaNet: A robust nested feature fusion coupled with attention incorporated width scaled lightweight architecture for Bengali sign language recognition IsharaNet：一个强大的嵌套特征融合，结合了关注宽度缩放的孟加拉手语识别轻量级架构

IF 4.5

Array Pub Date : 2025-08-11 DOI: 10.1016/j.array.2025.100486

Md Hasib Al Muzdadid Haque Himel, Md. Al Mehedi Hasan

{"title":"IsharaNet: A robust nested feature fusion coupled with attention incorporated width scaled lightweight architecture for Bengali sign language recognition","authors":"Md Hasib Al Muzdadid Haque Himel, Md. Al Mehedi Hasan","doi":"10.1016/j.array.2025.100486","DOIUrl":"10.1016/j.array.2025.100486","url":null,"abstract":"<div><div>Communication between ordinary and speech-hearing impaired people who interact mostly via sign language is one of the most significant challenges nowadays. For any class of individuals who try, learning and communicating with sign language is a difficult endeavor. Research on sign language recognition regarding various languages has been a long-standing concern, and several automated systems that have been proposed as a consequence have not yet proved to be particularly effective for Bengali, which has a wide vocabulary, character set, and expressive techniques, making it one of the most difficult sign languages. In this paper, a lightweight deep neural network architecture (IsharaNet) is proposed that incorporates parallel convolutional operations in order to yield a width scaled architecture in which nested feature fusion coupled with attention is leveraged. To enhance the network’s speed, the architecture has featured additional dropout layers and ReLU activation function. To evaluate the performance of the proposed architecture, the four most recently available Bengali sign language datasets, BdSL47, BdSLW-11, Shongket, and KU-BdSL, were employed. The highest Accuracy, F1-score, and AUC score reached 99.85%, 99.85%, and 0.999 in recognizing Bengali sign numerals. In recognizing Bengali sign alphabet, the highest Accuracy, F1-score, and AUC score reached 99.77%, 99.77%, and 0.999. The proposed architecture recognized Bengali sign words with the highest Accuracy of 99.09%, F1-score of 99.09%, and AUC score of 0.999. By demonstrating superior performance than other methods, the experimental findings indicate that the proposed architecture can be considered for simple and automated Bengali sign language recognition system.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"27 ","pages":"Article 100486"},"PeriodicalIF":4.5,"publicationDate":"2025-08-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144828738","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

Prediction accuracy in maritime simulator training performance assessment with varying data frequency 变数据频率下海事模拟器训练性能评估的预测精度

IF 4.5

Array Pub Date : 2025-08-11 DOI: 10.1016/j.array.2025.100489

Ziaul Haque Munim , Fabian Kjeldsberg , Tae-Eun Kim , Morten Bustgaard

{"title":"Prediction accuracy in maritime simulator training performance assessment with varying data frequency","authors":"Ziaul Haque Munim , Fabian Kjeldsberg , Tae-Eun Kim , Morten Bustgaard","doi":"10.1016/j.array.2025.100489","DOIUrl":"10.1016/j.array.2025.100489","url":null,"abstract":"<div><div>This study investigates how varying data sampling frequencies affect the classification accuracy of Machine Learning (ML) models when predicting student performance in maritime simulator training. ML-driven performance prediction is an essential part of Predictive Learning Analytics (PLA). If acceptable prediction accuracy can be achieved by using lower frequency data with larger time intervals between recorded data points, valuable resources in terms of data storage, handling, and computational cost, can be potentially saved. This study utilizes simulator log data from navigation students performing a <em>Williamson Turn</em> in both Ballast and Loaded ship conditions on a desktop simulator. Data frequencies ranging from 01 to 09 second intervals are examined. Results are evaluated by Area Under the Curve (AUC), Accuracy, Log Loss, Precision, Recall, and F1 Scores. The eXtreme Gradient Boosted Trees, variants of Keras Residual Neural Network, and Light Gradient Boosted Trees are trained on 87.5 % and tested on 12.5 % of the data. The best accuracy measurement scores are achieved on the 1-s frequency intervals in both ballast and loaded condition analysis. Further, the 1-s frequency intervals models are also the fastest and require less Random Access Memory (RAM). With reducing data frequency intervals, the model evaluation metrics deteriorate.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"27 ","pages":"Article 100489"},"PeriodicalIF":4.5,"publicationDate":"2025-08-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144912866","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

DiffViT-IBFD: A rolling bearing fault diagnosis approach based on diffusion model and vision transformer under data imbalance conditions DiffViT-IBFD：基于扩散模型和视觉变压器的数据不平衡条件下滚动轴承故障诊断方法

IF 4.5

Array Pub Date : 2025-08-07 DOI: 10.1016/j.array.2025.100483

Zheru Dong , Wen Zhao , Di Zhu , Zixin Zhang , Yuheng Ren

{"title":"DiffViT-IBFD: A rolling bearing fault diagnosis approach based on diffusion model and vision transformer under data imbalance conditions","authors":"Zheru Dong , Wen Zhao , Di Zhu , Zixin Zhang , Yuheng Ren","doi":"10.1016/j.array.2025.100483","DOIUrl":"10.1016/j.array.2025.100483","url":null,"abstract":"<div><div>Rolling Bearing fault data collected from industrial sites often exhibit class distribution imbalance, which significantly degrades the performance of deep learning-based intelligent bearing fault diagnosis models. Currently, most existing studies use the Generative Adversarial Network (GAN) to generate samples from the minority class, thereby improving the model's performance. However, the training process of GAN is highly unstable and susceptible to mode collapse, resulting in poor-quality and low-diversity generated samples. Given that the diffusion model was initially designed for image generation and its training process is relatively stable, a new rolling bearing fault diagnosis approach (DiffViT-IBFD) based on the diffusion model and Vision Transformer under data imbalance conditions is proposed. First, the proposed DiffViT-IBFD converts one-dimensional vibration data into two-dimensional time-frequency images through the short-time fourier transform. Second, the diffusion model typically adopts Unet as the backbone, which may cause gradient vanishing or explosion. Moreover, the skip connection strategy of Unet may effectively fail to integrate the low-level and high-level features of the data. An Unet-based ReC-Unet network is constructed to accurately and comprehensively extract data features, thereby enabling the diffusion model to generate higher-quality time-frequency samples. Finally, the patching operation in the Vision Transformer may lose the horizontal information in the time-frequency image. A horizontal slicing strategy (Hs-Patch) is developed to comprehensively extract the horizontal features of time-frequency images, thereby enhancing the feature expression capability of Vision Transformer. Experimental results on two publicly available datasets show that DiffViT-IBFD outperforms existing methods under data imbalance conditions, validating its effectiveness.</div></div>","PeriodicalId":8417,"journal":{"name":"Array","volume":"27 ","pages":"Article 100483"},"PeriodicalIF":4.5,"publicationDate":"2025-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144827572","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

Array最新文献