检索结果-内蒙古大学图书馆

HMM-Based Secure Framework for Optical Fog Devices in the Optical Fog/Cloud Network

Journal of Optical Communications 2023年第4期44卷 475-483页

作者： Sood, Sandeep K. Deep Singh, Kiran Department of Computer Science & Engineering Khalsa College of Engineering & Technology Amritsar India Department of Computer Science & Engineering Guru Nanak Dev University Regional Campus Punjab Gurdaspur India

In the present optical fog/cloud computing environment, optical line terminals and optical network units are used as the most promising optical fog devices (OFDs). The inherent characteristics of fog computing provide certain granted privileges to the attacker to hack devices and make them malicious. Also, existing security solutions generate false alarms that affect the performance of the underlying network. In this paper, a secure framework is proposed that not only predicts the malicious OFDs but also reduces false alarms. Hidden Markov model and intrusion detection system are used to detect the malicious one by computing the probability of shifting (POS) and then shift it to the virtual honeypot which is kept hidden by deploying it at the optical fog layer. In addition, it also reduces the generation of false alarm and logs all malicious activities for further analysis. In the experiment section, Python is used to validate the proposed framework. Further, HMM is simulated and tested in the MATLAB to reduce the false alarm rate. Results show that the proposed framework effectively reduces the false alarms and detects the malicious one and then shifts it onto the virtual honeypot efficiently. © 2019 Walter de Gruyter GmbH, Berlin/Boston.

关键词： Fog

来源：评论

学校读者我要写书评

暂无评论

Fraud Detection in Banking using the Kaggle Credit Card Dataset and XGBoost Model 5

Fraud Detection in Banking using the Kaggle Credit Card Data...

引用

5th International Conference on IoT Based Control Networks and Intelligent Systems, ICICNIS 2024

作者： Ango, Rithika Masih, Raj Kumar Reddy, C Kishor Kumar Shuaib, Mohammed Monika Singh, T. Alam, Shadab Stanley College of Engineering & Technology for Women Department of Computer Science and Engineering Hyderabad India Jazan University College of Engineering and Computer Science Department of Computer Science Jazan Saudi Arabia

ISBN: (纸本)9798331518097

The threat posed by credit card fraud, and by extension, online banking, continues to grow with the convenience brought forth by online banking services. Many financial institutions and customers stand at great risk because of this reality. The purpose of the paper is to enhance the security of banking systems through advanced technologies - machine learning. The research employed the well-known dataset from Kaggle Credit Card, which contains numerous records of transactions labeled as either fraud or not, thus presenting an aspect similar to real life imbalanced data. The main focus in the present work is the XGBoost model which involves one of the ensemble learning methods, which is characterized by a high level of precision and fast execution times. As a result of hyperparameter optimization and feature engineering, the paper also works on the performance of the algorithm improving its ability to detect fraudulent transactions in a sea of normal transactions. The results obtained from the experiments show that the XGBoost model is not only proven to be more accurate than earlier approaches to fraud detection, but also greatly reduces the number of false positives. This study not only contributes to the existing literature on detecting financial fraud but also represents the ways in which finance can benefit from machine learning to design secure systems within banking operations. The metrics in which the model suggested outperformed the available approaches are, the accuracy is 98.60%, error rate is 2.10% while its precision reached 95.40, recall of 92.30, F1 score 93.20 and AUC-ROC 99.70. This study not only contributes to the existing literature on detecting financial fraud but also represents the ways in which finance can benefit from machine learning to design secure systems within banking operations. © 2024 IEEE.

关键词： Banking

来源：评论

学校读者我要写书评

暂无评论

A Comprehensive Analysis of Feature Selection Techniques and Their Impact on Earthquake Damage Prediction

引用

Iranian Journal of science and Technology - Transactions of Civil engineering 2025年 1-35页

作者： Feroz, Shejuti Binte Sharmin, Nusrat Department of Computer Science and Engineering Military Institute of Science and Technology Dhaka Bangladesh

Earthquake damage prediction is vital to ensure occupants of buildings are not injured and substantial financial losses can be avoided. Algorithms based on machine learning are prevalent in this field. This study conducts an in depth analysis of the impact of feature selection on machine learning models for earthquake damage prediction. A feature selection method is critical for reducing data dimensionality, improving model accuracy, and mitigating overfitting. The Bat Algorithm, Particle Swarm Optimization, and Simulated Annealing were evaluated alongside twelve traditional feature selection techniques, such as Information Gain, Chi-square Test, Fisher’s Score, and LASSO Regularization. A comprehensive dataset from the 2015 Gorkha earthquake in Nepal was used to assess the predictive performance of three machine learning models-Decision Tree, Random Forest, and LightGBM. The novelty of this work lies in applying a broad spectrum of feature selection methods—many of which are being used for the first time—on the real-world dataset from the 2015 Gorkha earthquake in Nepal. A significant role is played by feature selection in the improvement of model accuracy and efficiency. Out of all the models, the Forward Feature Selection and Correlation Coefficient showed the best trade-off between processing time and accuracy. This study contributes to future disaster preparedness and structural resilience initiatives by highlighting the significance of meticulous feature selection and providing a useful framework for improving earthquake damage prediction models. © The Author(s), under exclusive licence to Shiraz University 2025.

关键词： Deep learning Earthquake prediction Machine learning

来源：评论

学校读者我要写书评

暂无评论

Security Challenges of Intent-Based Networking

引用

Communications of the ACM 2024年第7期67卷 56-65页

作者： Kim, Jiwon Okhravi, Hamed Tian, Dave Jing Ujcich, Benjamin E. Department of Computer Science Purdue University West LafayetteIN United States MIT Lincoln Laboratory LexingtonMA United States Department of Computer Science Georgetown University WashingtonDC United States

computer SYSTEMS HAVE evolved over decades to enable more flexible programmability. Unsurprisingly, such programmability has converged more closely to how humans think and speak. This is perhaps best exemplified in the evolution of programming languages: an obtuse set of ones and zeros (machine language), small snippets of instructions (assembly language), early low-level abstractions with human-readable syntax (for example, ALGOL or C), and today’s high-level abstractions with complex operations (for example, Python or Go). With each subsequent generation, programming languages have described richer semantics that enables desired functionality to be expressed more naturally. Not only has this evolution made programming easier, but such evolution allows humans to focus on what they want to do instead of how the machine should be programmed to do it. In other words, the evolution of high-level semantics has allowed humans to focus on the intent instead of the mechanism. ©2024 Copyright is held by the owner/author(s)

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Distributed Optimal Contract for Data Rewarding with Network Effects

引用

IEEE Transactions on Control of Network Systems 2024年第2期12卷 1332-1341页

作者： Baneshi, Alireza Montazeri, Mina Kebriaei, Hamed University of Tehran School of Electrical and Computer Engineering College of Engineering Tehran Iran Swiss Federal Laboratories for Materials Science and Technology Urban Energy Systems Laboratory Dübendorf Switzerland School of Computer Science Tehran Iran

Data rewarding is a novel business model that leads to an economic trend in mobile networks. In this scheme, the advertiser incentivizes mobile users (MUs) to watch advertisement (ads) and, in return, receive a reward in the form of mobile data. In this work, we model the interaction between an advertiser who has asymmetric information about MUs and MUs who are connected to each other under a network, using the contract theory approach. We obtain the necessary and sufficient conditions for an optimal and practical contract to motivate MUs to participate in the data rewarding scheme and encourage them to declare their private information truthfully. The formulation of this contract is a non-convex-constrained optimization problem. Using lemmas and propositions, we reformulate the initial optimization problem that is challenging to solve as an optimization problem with convex constraints and prove that these two problems are equivalent. Then, with the help of a distributed and non-convex algorithm, we obtain the amount of ads demand and incentive reward. © 2014 IEEE.

关键词： Contracts Optimization Social networking (online) Data models Network systems Control systems Indexes Distributed algorithms Couplings Complexity theory

来源：评论

学校读者我要写书评

暂无评论

SAM-Aware Graph Prompt Reasoning Network for Cross-Domain Few-Shot Segmentation 39

SAM-Aware Graph Prompt Reasoning Network for Cross-Domain Fe...

引用

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Peng, Shi-Feng Sun, Guolei Li, Yong Wang, Hongsong Xie, Guo-Sen School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing China Computer Vision Laboratory ETH Zurich Switzerland School of Computer Science and Engineering Southeast University Nanjing China

ISBN: (纸本)157735897X

The primary challenge of cross-domain few-shot segmentation (CD-FSS) is the domain disparity between the training and inference phases, which can exist in either the input data or the target classes. Previous models struggle to learn feature representations that generalize to various unknown domains from limited training domain samples. In contrast, the large-scale visual model SAM, pre-trained on tens of millions of images from various domains and classes, possesses excellent generalizability. In this work, we propose a SAM-aware graph prompt reasoning network (GPRN) that fully leverages SAM to guide CD-FSS feature representation learning and improve prediction accuracy. Specifically, we propose a SAM-aware prompt initialization module (SPI) to transform the masks generated by SAM into visual prompts enriched with high-level semantic information. Since SAM tends to divide an object into many sub-regions, this may lead to visual prompts representing the same semantic object having inconsistent or fragmented features. We further propose a graph prompt reasoning (GPR) module that constructs a graph among visual prompts to reason about their interrelationships and enable each visual prompt to aggregate information from similar prompts, thus achieving global semantic consistency. Subsequently, each visual prompt embeds its semantic information into the corresponding mask region to assist in feature representation learning. To refine the segmentation mask during testing, we also design a non-parameter adaptive point selection module (APS) to select representative point prompts from query predictions and feed them back to SAM to refine inaccurate segmentation results. Experiments on four standard CD-FSS datasets demonstrate that our method establishes new state-of-the-art results. Copyright © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Oppositional Red Fox Optimization Based Task Scheduling Scheme for Cloud Environment

引用

computer Systems science & engineering 2023年第4期45卷 483-495页

作者： B.Chellapraba D.Manohari K.Periyakaruppan M.S.Kavitha Department of Information Technology Karpagam Institute of TechnologyCoimbatore641032TamilnaduIndia Department of Computer Science and Engineering St.Joseph’s Institute of TechnologyChennai600119India Department of Computer Science&Engineering SNS College of EngineeringCoimbatore641107India Department of Computer Science&Engineering SNS College of TechnologyCoimbatore641035India

Owing to massive technological developments in Internet of Things(IoT)and cloud environment,cloud computing(CC)offers a highly flexible heterogeneous resource pool over the network,and clients could exploit various resources on *** IoT-enabled models are restricted to resources and require crisp response,minimum latency,and maximum bandwidth,which are outside the *** was handled as a resource-rich solution to aforementioned *** high delay reduces the performance of the IoT enabled cloud platform,efficient utilization of task scheduling(TS)reduces the energy usage of the cloud infrastructure and increases the income of service provider via minimizing processing time of user ***,this article concentration on the design of an oppositional red fox optimization based task scheduling scheme(ORFOTSS)for IoT enabled cloud *** presented ORFO-TSS model resolves the problem of allocating resources from the IoT based cloud *** achieves the makespan by performing optimum TS procedures with various aspects of incoming *** designing of ORFO-TSS method includes the idea of oppositional based learning(OBL)as to traditional RFO approach in enhancing their efficiency.A wide-ranging experimental analysis was applied on the CloudSim *** experimental outcome highlighted the efficacy of the ORFO-TSS technique over existing approaches.

关键词： Metaheuristics task scheduling cloud computing internet of things makespan red fox optimizer

来源：评论

学校读者我要写书评

暂无评论

Lung Cancer Detection by Employing Adaptive Entropy Variance Dropout Regularization in GAN Variants

引用

SN computer science 2024年第3期5卷 325页

作者： Thirumagal, E. Saruladha, K. Department of Computer Science and Engineering Puducherry Technological University Puducherry India School of Computer Science and Engineering REVA University Bengaluru India

Lung cancer segmentation using Deep Neural Networks (DNN) needs accurate pixel-level data which is typically small. This leads to overfitting issue, and in order to alleviate this, the research is been done on L2 regularization and dropout algorithms which are commonly used in DNNs. To solve this co-adaptation problem, this paper discusses the basics of existing L2 regularization and dropout algorithms and their limitations. Further, this paper proposes a novel entropy variance dropout regularization algorithm to alleviate the existing limitations and enhance the model accuracy. The proposed regularization technique drops the co-adaptation features using the information entropy and variance. The experiments were conducted on Generative Adversarial Networks (GAN) variants using L2 regularization, dropout regularization and proposed Adaptive Entropy Variance (AEV) dropout regularization. The experimental results prove that the GAN variants using the proposed method alleviate the overfitting problem and also yield high accuracy. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd 2024.

关键词： Dropout Entropy GAN Lung cancer Regularization Variance

来源：评论

学校读者我要写书评

暂无评论

Adaptive pelican optimization with optimized mask RCNN for automatic lung cancer detection

引用

Multimedia Tools and Applications 2025年 1-21页

作者： Sudha, R. Maheswari, K.M. Uma Department of Computer Science and Engineering SRM Institute of Science and Technology Chengalpattu India Department of Computing Technologies SRM Institute of Science and Technology Chengalpattu India

Lung cancer is a prevalent and deadly disease worldwide, necessitating accurate and timely detection methods for effective treatment. Deep learning-based approaches have emerged as promising solutions for automated medical image analysis. This study proposes an enhanced Mask R-CNN framework tailored specifically for the automatic detection and severity analysis of lung cancer from CT images. The proposed approach consists of three stages namely, pre-processing, lung nodule detection, and segmentation using enhanced Mask R-CNN and severity analysis. Our framework employs a deep convolutional neural network architecture trained on a comprehensive dataset of annotated lung images. By incorporating a region-based convolutional neural network (R-CNN) with a mask prediction branch, our model accurately localizes lung tumors while providing precise pixel-level segmentation masks. To enhance the performance of mask RCNN, the parameter present in the classifier is optimally selected using the adaptive pelican optimization (APO) algorithm. The proposed framework detects lung tumors and provides a comprehensive severity analysis, enabling clinicians to assess cancer stage and progression accurately. Evaluation on a benchmark dataset demonstrates superior detection accuracy and robustness compared to existing methods. Our enhanced Mask R-CNN approach shows promise as a valuable tool for early diagnosis and severity assessment of lung cancer, potentially improving patient outcomes and healthcare efficiency. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Lung cancer

来源：评论

学校读者我要写书评

暂无评论

Failure Prediction for Scientific Workflows Using Nature-Inspired Machine Learning Approach

引用

Intelligent Automation & Soft Computing 2023年第4期36卷 223-233页

作者： S.Sridevi Jeevaa Katiravan Department of Computer Science and Engineering Velammal Engineering CollegeChennai600066India

Scientiﬁc workﬂows have gained the emerging attention in sophisti-cated large-scale scientiﬁc problem-solving *** pay-per-use model of cloud,its scalability and dynamic deployment enables it suited for executing scientiﬁc workﬂow *** the cloud is not a utopian environment,failures are inevitable that may result in experiencingﬂuctuations in the delivered *** a single task failure occurs in workﬂow based applications,due to its task dependency nature,the reliability of the overall system will be affected *** rather than reactive fault-tolerant approaches,proactive measures are vital in scientiﬁc workﬂ*** work puts forth an attempt to con-centrate on the exploration issue of structuring a nature inspired metaheuristics-Intelligent Water Drops Algorithm(IWDA)combined with an efﬁcient machine learning approach-Support Vector Regression(SVR)for task failure prognostica-tion which facilitates proactive fault-tolerance in the scheduling of scientiﬁc workﬂow *** failure prediction models in this study have been implemented through SVR-based machine learning approaches and the precision accuracy of prediction is optimized by IWDA and several performance metrics were evaluated on various benchmark workﬂ*** experimental results prove that the proposed proactive fault-tolerant approach performs better compared with the other existing techniques.

关键词： Failure prediction intelligent water drops support vector regression proactive fault-tolerance scientiﬁc workﬂows precision accuracy resource provisioning

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：