检索结果-内蒙古大学图书馆

Hybrid Sampling Technique for Imbalanced Android Malware Family Classification

SN computer science 2025年第3期6卷 1-16页

作者： Chauhan, Kshamta Gandotra, Ekta Department of Computer Science and Engineering and Information Technology Jaypee University of Information Technology Waknaghat Himachal Pradesh Solan India

The rapid explosion of Android-based devices has led to a disturbing surge in the volume and sophistication of Android malware. Effective classification of these malicious applications is essential for safeguarding user security and privacy. However, the imbalanced nature of the Android malware data, where some of the families outnumber the other ones, poses a significant challenge to traditional machine learning algorithms. The issue of imbalanced data can be handled by oversampling and undersampling techniques. Oversampling raises the number of instances in the minority class and undersampling decreases the number of instances in the majority class in order to balance the class distribution. Oversampling and undersampling can lead to overfitting and loss of information. To combine the strengths of both techniques, hybrid sampling is used. This paper proposes hybrid sampling techniques that combine oversampling and undersampling methods. Two oversampling techniques, i.e., synthetic minority oversampling technique (SMOTE), and random oversampling (ROS), and three undersampling techniques, i.e., tomek's links (TOMEK), edited nearest neighbours (ENN), and neighborhood cleaning rule (NCR), are used. The proposed hybrid sampling methods are obtained after combining each oversampling technique with an undersampling one. All the hybrid sampling techniques are evaluated on a self-created malware dataset that consists of multiple imbalanced classes of malware families. Four machine learning models, i.e., support vector machine (SVM), decision tree (DT), random forest (RF), and k-nearest neighbours (K-NN), are trained after applying the hybrid sampling techniques. From the experimental results, it is concluded that ROS-NCR outperforms other methods in handling imbalanced Android malware families. It achieves 95.81% accuracy and an F-measure of 0.95 using RF. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2025.

关键词： Android malware Hybrid sampling Imbalanced data Machine learning Multi-class classification

来源：评论

学校读者我要写书评

暂无评论

Preventing 51% Attack by Using Consecutive Block Limits in Bitcoin

引用

IEEE Access 2024年 12卷 77852-77869页

作者： Mahmood Babur, Sohail Ur Rehman Khan, Shafiq Yang, Jing Chen, Yen-Lin Soon Ku, Chin Yee Por, Lip Government Murray College Department of Information Technology Sialkot51010 Pakistan Capital University of Science and Technology Department of Computer Science Islamabad44730 Pakistan Universiti Malaya Department of Computer System and Technology Kuala Lumpur50603 Malaysia National Taipei University of Technology Department of Computer Science and Information Engineering Taipei106344 Taiwan Universiti Tunku Abdul Rahman Department of Computer Science Kampar31900 Malaysia

In permissionless blockchain systems, Proof of Work (PoW) is utilized to address the issues of double-spending and transaction starvation. When an attacker acquires more than 50% of the hash power of the entire network, they gain the ability to engage in double-spending activities, posing a significant threat to the PoW consensus algorithm. This research focuses on the consensus algorithm employed in the Bitcoin system, explaining how it operates and the security challenges it faces. The proposed modification to the PoW algorithm imposes a restriction on miners: they are not allowed to accept consecutive blocks from the same miner into the final local blockchain to prevent the 51% attack problem. This modification supports transactions that require six confirmations. In the event an attacker attempts a 51% attack with a private chain that consists of fewer than 6 blocks, it becomes easier to detect a double-spending attack before accepting the attacker's private chain. The modified algorithm introduces a 'Safe Mode Detection Algorithm' that scrutinizes incoming blocks for adjustments at the top of the local blockchain. If inconsistencies are identified, the consensus algorithm proceeds cautiously by comparing the UTXO dictionaries from the attacker's chain with those from the miner's own blockchain. This meticulous comparison aims to detect instances of double-spending. If such instances are detected, the miner rejects the attacker's chain, establishing a double-spend-free environment and thwarting 51% attacks. © 2013 IEEE.

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

Robust tube-based MPC with smooth computation for dexterous robot manipulation

引用

science China(information sciences) 2024年第11期67卷 252-268页

作者： Yu LUO Tianying JI Fuchun SUN Qie SIMA Huaping LIU Mingxuan JING Jianwei ZHANG Department of Computer Science and Technology Tsinghua University Science & Technology on Integrated Information System Laboratory Institute of Software Chinese Academy of Sciences Department of Informatics University of Hamburg

Dexterous robot manipulation has shone in complex industrial scenarios, where multiple manipulators, or fingers, cooperate to grasp and manipulate objects. When encountering multi-objective optimization with system constraints in such scenarios, model predictive control(MPC) has demonstrated exceptional performance in complex multi-robot manipulation tasks involving multi-objective optimization with system constraints. However, in such scenarios, the substantial computational load required to solve the optimal control problem(OCP) at each triggering instant can lead to significant delays between state sampling and control application, hindering real-time performance. To address these challenges, this paper introduces a novel robust tube-based smooth MPC approach for two fundamental manipulation tasks: reaching a given target and tracking a reference trajectory. By predicting the successor state as the initial condition for imminent OCP solving, we can solve the forthcoming OCP ahead of time, alleviating delay effects. Additionally,we establish an upper bound for linearizing the original nonlinear system, reducing OCP complexity and enhancing response speed. Grounded in tube-based MPC theory, the recursive feasibility and closed-loop stability amidst constraints and disturbances are ensured. Empirical validation is provided through two numerical simulations and two real-world dexterous robot manipulation tasks, which shows that the seamless control input by our methods can effectively enhance the solving efficiency and control performance when compared to conventional time-triggered MPC strategies.

关键词： dexterous robot manipulation model predictive control smooth computation

来源：评论

学校读者我要写书评

暂无评论

A Deep Learning Approach for Landmines Detection Based on Airborne Magnetometry Imaging and Edge Computing

引用

computer Modeling in engineering & sciences 2024年第5期139卷 2117-2137页

作者： Ahmed Barnawi Krishan Kumar Neeraj Kumar Bander Alzahrani Amal Almansour Faculty of Computing and Information Technology King Abdulaziz UniversityJeddahSaudi Arabia Department of Computer Science and Engineering Thapar Institute of Engineering and TechnologyDeemed to be UniversityPatialaIndia

Landmines continue to pose an ongoing threat in various regions around the world,with countless buried landmines affecting numerous human *** detonation of these landmines results in thousands of casualties reported worldwide ***,there is a pressing need to employ diverse landmine detection techniques for their *** effective approach for landmine detection is UAV(Unmanned Aerial Vehicle)based AirborneMagnetometry,which identifies magnetic anomalies in the local terrestrial magnetic *** can generate a contour plot or heat map that visually represents the magnetic field *** the effectiveness of this approach,landmine removal remains a challenging and resource-intensive task,fraughtwith *** computing,on the other hand,can play a crucial role in critical drone monitoring applications like landmine *** processing data locally on a nearby edge server,edge computing can reduce communication latency and bandwidth requirements,allowing real-time analysis of magnetic field *** enables faster decision-making and more efficient landmine detection,potentially saving lives and minimizing the risks involved in the ***,edge computing can provide enhanced security and privacy by keeping sensitive data close to the source,reducing the chances of data exposure during *** paper introduces the MAGnetometry Imaging based Classification System(MAGICS),a fully automated UAV-based system designed for landmine and buried object detection and *** have developed an efficient deep learning-based strategy for automatic image classification using magnetometry dataset *** simulating the proposal in various network scenarios,we have successfully detected landmine signatures present in themagnetometry *** trained models exhibit significant performance improvements,achieving a maximum mean average precision value of 97.8%.

关键词： CNN deep learning landmine detection magnetometer mean average precision UAV

来源：评论

学校读者我要写书评

暂无评论

A multi-fused convolutional neural network model for fruit image classification

引用

International Journal of Cognitive Computing in engineering 2024年第1期5卷 416-424页

作者： Sinha, Bam Bahadur Dhanalakshmi, R. Computer Science and Engineering National Institute of Technology Sikkim 737139 India Computer Science and Engineering Indian Institute of Information Technology Tiruchirappalli Tamil Nadu 620012 India

Fruit categorization presents a significant challenge due to the diverse range of fruit types and their similarities in color, shape, size, and structure. This challenge is addressed in this research by proposing a multi-fused CNN approach that integrates three state-of-the-art deep learning models, namely EfficientNet-B0, MobileNetV2, and ResNet50V2, into a unified pipeline. The primary objective is to enhance prediction accuracy and mitigate overfitting issues commonly encountered with conventional techniques. Utilizing the Fruit-360 dataset (both small and full versions), the proposed multi-fused CNN model demonstrates superior performance, achieving an impressive accuracy rate of 99.32% and 97.15% in classifying fruit types across 24 and 131 different categories, respectively. This substantial improvement underscores the potential of the proposed approach in reliably detecting fruit types. In future work, efforts will be directed towards optimizing the model for enhanced processing speed, facilitating real-time performance in practical applications, particularly in agricultural automation. Additionally, rigorous experiments on diverse real-world fruit images will be conducted to further validate the robustness of the proposed model. © 2024 The Authors

关键词： Fruits

来源：评论

学校读者我要写书评

暂无评论

Gradient Propagation Based DenseNet121 with ResNet50 Feature Extraction for Lymphoma Classification

引用

Journal of The Institution of Engineers (India): Series B 2024年 1-13页

作者： Srinivasan, Deepthi Kalaiarasan, C. Department of Computer Science and Engineering Presidency University Karnataka Bengaluru India Department of Computer Science and Engineering and Information Science Presidency University Karnataka Bengaluru India

Lymphoma is a type of malignant tumor that develops from lymphoid hematopoietic tissues. The precise diagnosis of lymphomas is one of the challenging tasks because of the similarity within the morphological features across lymphoma classes. Therefore, an effective classification of lymphoma is required to be developed for taking a timely action for the patients. This paper aims to propose the Gradient Propagation based DenseNet121 namely GPDN for an effective multiclass lymphoma classification. The dense connections of GPDN is used to mitigate the vanishing gradient issue by confirming an effective gradient propagation over the network for enhancing stability and avoiding overfitting issue. An effective reutilization of features from previous layers, the GPDN obtains the intricate for enhanced differentiation of lymphoma. Different pre-processing approaches such as gaussian filter, min-max normalization and data augmentation are developed for denoising, rescaling and augmenting the input images. Here, multiscale features from the image are obtained by using ResNet50-based feature extraction. This research uses a dataset from Kaggle named Multi cancer dataset. The performance of the proposed ResNet50-GPDN is analyzed using accuracy, precision, recall and F-measure. Simulation results show that the GPDN achieves higher accuracy of 99.90% than the VGG16, InceptionV3, MobileNet and AlexNet. Moreover, existing researches HPC and FFNN-ResNet50, are used for the comparison of the ResNet50-GPDN. The accuracy of ResNet50-GPDN is 99.90% which is clearly higher than those of HPC and FFNN-ResNet50. © The Institution of Engineers (India) 2024.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Autonomous embodied navigation task generation from natural language dialogues

引用

science China(information sciences) 2025年第5期68卷 119-132页

作者： Haifeng XU Yongchang LI Lumeng MA Chunwen LI Yanzhi DONG Xiaohu YUAN Huaping LIU Department of Automation Tsinghua University School of Physics and Electronic Information Yantai University School of Electrical and Electronic Engineering Shanghai Institute of Technology Department of Computer Science and Technology Tsinghua University

Robots are increasingly being deployed in densely populated environments, such as homes, hotels, and office buildings, where they rely on explicit instructions from humans to perform tasks. However, complex tasks often require multiple instructions and prolonged monitoring, which can be time-consuming and demanding for users. Despite this, there is limited research on enabling robots to autonomously generate tasks based on real-life scenarios. Advanced intelligence necessitates robots to autonomously observe and analyze their environment and then generate tasks autonomously to fulfill human requirements without explicit commands. To address this gap, we propose the autonomous generation of navigation tasks using natural language dialogues. Specifically, a robot autonomously generates tasks by analyzing dialogues involving multiple persons in a real office environment to facilitate the completion of item transportation between various *** propose the leveraging of a large language model(LLM) through chain-of-thought prompting to generate a navigation sequence for a robot from dialogues. We also construct a benchmark dataset consisting of 625 multiperson dialogues using the generation capability of LLMs. Evaluation results and real-world experiments in an office building demonstrate the effectiveness of the proposed method.

关键词： proactive robot robot navigation service robot large language model

来源：评论

学校读者我要写书评

暂无评论

Adversarial attacks and defenses for digital communication signals identification

引用

Digital Communications and Networks 2024年第3期10卷 756-764页

作者： Qiao Tian Sicheng Zhang Shiwen Mao Yun Lin College of Computer Science and Technology Harbin Engineering UniversityHarbin150001China College of Information and Communication Engineering Harbin Engineering UniversityHarbin150000China Department of Electrical and Computer Engineering Auburn UniversityAuburnAL36849USA

As modern communication technology advances apace,the digital communication signals identification plays an important role in cognitive radio networks,the communication monitoring and management *** has become a promising solution to this problem due to its powerful modeling capability,which has become a consensus in academia and ***,because of the data-dependence and inexplicability of AI models and the openness of electromagnetic space,the physical layer digital communication signals identification model is threatened by adversarial *** examples pose a common threat to AI models,where well-designed and slight perturbations added to input data can cause wrong ***,the security of AI models for the digital communication signals identification is the premise of its efficient and credible *** this paper,we first launch adversarial attacks on the end-to-end AI model for automatic modulation classifi-cation,and then we explain and present three defense mechanisms based on the adversarial *** we present more detailed adversarial indicators to evaluate attack and defense ***,a demonstration verification system is developed to show that the adversarial attack is a real threat to the digital communication signals identification model,which should be paid more attention in future research.

关键词： Digital communication signals identification AI model Adversarial attacks Adversarial defenses Adversarial indicators

来源：评论

学校读者我要写书评

暂无评论

AI-enabled dental caries detection using transfer learning and gradient-based class activation mapping

引用

Journal of Ambient Intelligence and Humanized Computing 2024年第7期15卷 3009-3033页

作者： Inani, Hardik Mehta, Veerangi Bhavsar, Drashti Gupta, Rajeev Kumar Jain, Arti Akhtar, Zahid Department of Computer Science and Engineering Pandit Deendayal Energy University Gujarat Gandhinagar India Department of Computer Science and Engineering and Information Technology Jaypee Institute of Information Technology Uttar Pradesh Noida India Department of Network and Computer Security State University of New York Polytechnic Institute NY United States

Dental caries detection holds the key to unlocking brighter smiles and healthier lives by identifying one of the most common oral health issues early on. This vital topic sheds light on innovative ways to combat tooth decay, empowering individuals to take control of their oral health and maintain radiant smiles. This research paper delves into the realm of transfer learning techniques, aiming to elevate the precision and efficacy of dental caries diagnosis. Utilizing Keras ImageDataGenerator, a rich and balanced dataset is crafted by augmenting teeth images from the Kaggle teeth dataset. Five cutting-edge pre-trained architectures are harnessed in the transfer learning approach: EfficientNetV2B3, VGG19, InceptionResNetV2, Xception, and ResNet50, with each model, initialized using ImageNet weights and tailored top layers. A comprehensive set of evaluation metrics, encompassing accuracy, precision, recall, F1-score, and false negative rates are employed to gauge the performance of these architectures. The findings unveil the unique advantages and drawbacks of each model, illuminating the path to an optimal choice for dental caries detection using Grad-CAM (Gradient-weighted Class Activation Mapping). The testing accuracies achieved by EfficientNetV2B3, VGG19, InceptionResNetV2, Xception, and ResNet50 models stand at 95.89%, 96.58%, 93.15%, 93.15%, and 94.18%, respectively. The Training accuracies stood at 100%, 99.91%, 100%, 100% and 100%, meanwhile on validation we achieved 97.63%, 96.68%, 98.82%, 96.68%, and 100% accuracies for EfficientNetV2B3, VGG19, InceptionResNetV2, Xception, and ResNet50 models respectively. Capitalizing on transfer learning and juxtaposing diverse pre-trained architectures, this research paper paves the way for substantial advancements in dental diagnostic capabilities, culminating in enhanced patient outcomes and superior oral health. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2024.

关键词： Chemical activation

来源：评论

学校读者我要写书评

暂无评论

Hybrid CNN-ViT architecture to exploit spatio-temporal feature for fire recognition trained through transfer learning

引用

Multimedia Tools and Applications 2025年第8期84卷 4703-4732页

作者： Shahid, Mohammad Wang, Hong-Cyuan Chen, Yung-Yao Hua, Kai-Lung Dept. of Computer Science and Information Engineering National Taiwan University of Science and Technology Taipei Taiwan Dept. of Electronic and Computer Engineering National Taiwan University of Science and Technology Taipei Taiwan

Fires are becoming one of the major natural hazards that threaten the ecology, economy, human life and even more worldwide. Therefore, early fire detection systems are crucial to prevent fires from spreading out of control and causing destruction. Based on vision sensors, many fire detection techniques have evolved with the recent surge of curiosity in deep learning, which exploits the spatial features of individual images. However, fire can take different forms, scales, and combustion materials can produce different colors, making accurate fire detection from an image challenging. Small fires captured from long-distance cameras lack salient features, further complicating detection. This paper proposes a hybrid structure that uses attention-enhanced convolutional neural networks and vision transformers (CNN-ViT) to detect fire. The proposed CNN-ViT first pays spatial attention to every frame and then aggregates temporal contextual information from neighboring frames to improve detection performance. Due to the limited availability of training fire datasets, the study employs deep transfer learning for feature extraction using pre-trained CNN. We used various metrics to examine the efficacy of the proposed approach. The results showed that the CNN-ViT method outperformed previous models based on spatial-temporal characteristics by achieving a relative improvement in accuracy and F1 score. The satisfactory results on images contaminated with different intensities of noise confirm the robustness of the approach. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Fires

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：