检索结果-内蒙古大学图书馆

Modeling of Optimal Deep Learning Based Flood Forecasting Model Using Twitter Data

Intelligent Automation & Soft Computing 2023年第2期35卷 1455-1470页

作者： G.Indra N.Duraipandian Department of Computer Science and Engineering RMK College of Engineering and TechnologyChennai601206India Department of Computer Science and Engineering Saveetha Engineering CollegeChennai602106India

Aﬂood is a signiﬁcant damaging natural calamity that causes loss of life and *** work on the construction ofﬂood prediction models intended to reduce risks,suggest policies,reduce mortality,and limit property damage caused byﬂ*** massive amount of data generated by social media platforms such as Twitter opens the door toﬂood *** of the real-time nature of Twitter data,some government agencies and authorities have used it to track natural catastrophe events in order to build a more rapid rescue ***,due to the shorter duration of Tweets,it is difﬁcult to construct a perfect prediction model for determiningﬂ*** learning(ML)and deep learning(DL)approaches can be used to statistically developﬂood prediction *** the same time,the vast amount of Tweets necessitates the use of a big data analytics(BDA)tool forﬂood *** this regard,this work provides an optimal deep learning-basedﬂood forecasting model with big data analytics(ODLFF-BDA)based on Twitter *** suggested ODLFF-BDA technique intends to anticipate the existence ofﬂoods using tweets in a big data *** ODLFF-BDA technique comprises data pre-processing to convert the input tweets into a usable *** addition,a Bidirectional Encoder Representations from Transformers(BERT)model is used to generate emotive contextual embed-ding from ***,a gated recurrent unit(GRU)with a Multilayer Convolutional Neural Network(MLCNN)is used to extract local data and predict theﬂ***,an Equilibrium Optimizer(EO)is used toﬁne-tune the hyper-parameters of the GRU and MLCNN models in order to increase prediction *** memory usage is pull down lesser than 3.5 MB,if its compared with the other algorithm *** ODLFF-BDA technique’s performance was validated using a benchmark Kaggle dataset,and theﬁndings showed that it outperformed other recent approaches signiﬁcantly.

关键词： Big data analytics predictive models deep learning ﬂood prediction twitter data hyperparameter tuning

来源：评论

学校读者我要写书评

暂无评论

Optimizing Clustering in Wireless Sensor Networks: A Synergistic Approach Using Reinforcement Learning (RL) and Particle Swarm Optimization (PSO)

引用

SN computer science 2024年第6期5卷 718页

作者： Choudhary, Arpita Barwar, N.C. Computer Science and Engineering Department MBM University

Significant research efforts are currently devoted to wireless sensor networks due to its broad range of applications. WSNs face various constraints, encompassing challenges related to communication, clustering management and the finite battery life of nodes. Thus, Energy conservation in such networks is indispensable. Given a constant energy consumption rate during information sensing and reception, the highest energy consumption among sensor nodes occurs during data transmission. One of promising solution to reduce energy consumption is organizing WSN in clusters. Clustering in Wireless Sensor Networks (WSN) involves grouping sensor nodes into clusters to facilitate efficient data aggregation, communication, and management within the network. This organizational structure helps optimize energy consumption, enhance scalability, and prolong the overall lifespan of the WSN. However determining the optimal criteria for selecting cluster heads is challenging, as it involves balancing energy efficiency, network connectivity, and load distribution. In this paper, a dual-phase approach is proposed, firstly Reinforcement learning (RL) approach has been applied to clustering in WSNs which enables nodes to autonomously adapt their clustering strategies, leading to more efficient and adaptive network configurations. Further Particle Swarm Optimization (PSO) can be utilized for cluster head selection in Wireless Sensor Networks (WSNs) to optimize the formation of clusters. The consideration of both local and global perspectives in the proposed approach results in a more balanced and efficient clustering solution. The outcomes of our experiments demonstrate the enhanced performance of the integrated approach as compared to traditional clustering algorithms. Results show considerable improvement in terms of reduced energy consumption, accuracy and efficiency in fault detection specifically tailored for Wireless Sensor Networks (WSNs). In addition the proposed algorithm show enha

关键词： Clustering Energy-saving mechanisms Optimization techniques Particle Swarm Optimization Reinforcement learning Wireless Sensor Networks

来源：评论

学校读者我要写书评

暂无评论

Property-guided generation of complex polymer topologies using variational autoencoders

引用

npj Computational Materials 2024年第1期10卷 1825-1837页

作者： Shengli Jiang Adji Bousso Dieng Michael A.Webb Department of Chemical and Biological Engineering Princeton UniversityPrincetonNJUSA Vertaix Department of Computer SciencePrinceton UniversityPrincetonNJUSA

The complexity and diversity of polymer topologies,or chain architectures,present substantial challenges in predicting and engineering polymer *** machine learning is increasingly used in polymer science,applications to address architecturally complex polymers are ***,we use a generative machine learning model based on variational autoencoders and data generated from molecular dynamics simulations to design polymer topologies that exhibit target properties.

关键词： properties variational polymer

来源：评论

学校读者我要写书评

暂无评论

VoteDroid: a new ensemble voting classifier for malware detection based on fine-tuned deep learning models

引用

Multimedia Tools and Applications 2025年第12期84卷 10923-10944页

作者： Bakır, Halit Faculty of Engineering and Natural Sciences Department of Computer Engineering Sivas University of Science and Technology Sivas Turkey

In this work, VoteDroid a novel fine-tuned deep learning models-based ensemble voting classifier has been proposed for detecting malicious behavior in Android applications. To this end, we proposed adopting the random search optimization algorithm for deciding the structure of the models used as voter classifiers in the ensemble classifier. We specified the potential components that can be used in each model and left the random search algorithm taking a decision about the structure of the model including the number of each component that should be used and its location in the structure. This optimization method has been used to build three different deep learning models namely CNN-ANN, pure CNN, and pure ANN. After selecting the best structure for each DL model, the selected three models have been trained and tested using the constructed image dataset. Afterward, we suggested hybridizing the fine-tuned three deep-learning models to form one ensemble voting classifier with two different working modes namely MMR (Malware Minority Rule) and LMR (Label Majority Rule). To our knowledge, this is the first time that an ensemble classifier has been fine-tuned and hybridized in this way for malware detection. The results showed that the proposed models were promising, where the classification accuracy exceeded 97% in all experiments. © The Author(s) 2024.

关键词： Android malware

来源：评论

学校读者我要写书评

暂无评论

Towards Accurate Renal Micro-Structure Segmentation: An Assessment of YOLOv8 and Mask R-CNN Models

Towards Accurate Renal Micro-Structure Segmentation: An Asse...

引用

2024 IEEE International Women in engineering (WIE) Conference on electrical and computer engineering, WIECON-ECE 2024

作者： Ahmed, Md. Estiak Pinki, Farhana Tazmim Quader, Mozdaher Abdul Quader, Montaser Abdul Ayrin, Fateha Jannat Anwar, Ahmed Selim Roy, Prosenjit Computer Science and Engineering Discipline Bangladesh Department of Computer Science and Engineering Bangladesh Department of Computer Science and Engineering Bangladesh Department of Eee Bangladesh Prairie View A & M University Department of Computer Science TX United States

ISBN: (纸本)9798331535476

In the field of medical imaging, correct instance segmentation is essential. This work attempts to address the problems related to renal micro-structure segmentation by using the power of YOLOv8 and special MASK R-CNN models to facilitate the development of more rapid and accurate nephrology diagnosis, which will benefit physicians and other healthcare providers. Improved medical imaging analysis is made possible by the model's high precision rate of 'glomerulus' and 'blood vessel' structure detection. The application of the MASKRCNN framework, which successfully integrated pertinent annotations to boost visual understanding, increased the clarity of image interpretation. Additionally, this work uses the HuBMAP dataset to segregate renal microstructures using the YOLOv8 and Mask R-CNN techniques. Strong performance was shown by the YOLOv8 model, which received an IoU score of 84.55%. The Mask R-CNN model received an IoU score of 80.66%, indicating its potential. It is an impressive accomplishment to obtain a high IoU score for YOLOv8 in the context of medical photo segmentation, since this suggests that there is still room for further improvement and optimization of the model. For medical professionals, the IoU scores are crucial since they offer a technique to improve accuracy and reliability when it comes to nephrology diagnosis. In order to improve diagnosis accuracy and enable broader clinical adoption, future research aims to improve these IoU ratings. The observed disparity underscored the inherent complexities associated with differentiating between ambiguous kidney structures and also the pressing need for further advancement. Our preliminary work demonstrates the potential of MASK-RCNN and YOLOv8 in the challenging field of renal micro-structure segmentation. © 2024 IEEE.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Triplet-Style Dynamic Graph Network With Transformer Encoder for Scam Detection in Cryptocurrency Transactions

引用

IEEE Access 2025年 13卷 94402-94415页

作者： Nam, Min-Woo Lee, Hyeon-Ju Buu, Seok-Jun Gyeongsang National University Department of Computer Science and Engineering Gyeongsangnam-do Jinju52828 Korea Republic of Gyeongsang National University School of Aerospace Engineering Department of Computer Science and Engineering Gyeongsangnam-do Jinju52828 Korea Republic of

The surge in cryptocurrencies has been accompanied by a significant rise in scams, underscoring the critical need for precise scam detection. Cryptocurrency markets and transaction networks are dynamic, leading to evolving scam tactics and transaction topologies that challenge detection efforts. Compounding this, scammers cleverly mimic benign transactions, blurring the lines between legitimate and illicit activities. To address these intertwined challenges, we introduce the Triplet-style Dynamic Graph Convolutional Network (TD-GCN). TD-GCN is specifically engineered to model the dynamic nature of cryptocurrency transaction networks and to disentangle subtle distinctions between scam and benign transaction patterns. It leverages a Transformer Encoder for dynamic weight updates, effectively capturing network evolution, and employs Triplet Learning to disentangle representations of similar-appearing scam and benign transactions. Evaluations on a real-world Bitcoin transaction network demonstrate TD-GCN’s superior scam detection performance. Notably, TD-GCN achieves a leading F1-macro score of 0.8855 and a 10.36%p precision increase over competing models, crucial for minimizing false positives. Ablation studies highlight the Triplet Dynamic Disentangle component’s key role, reducing false positives by 81.1%. TD-GCN’s performance gains stem from its dynamic updates for evolving networks and Triplet Learning for disentangling subtly different patterns. These features enable TD-GCN to significantly bolster cryptocurrency security by effectively detecting scams and minimizing false positives in dynamic transaction networks. © 2013 IEEE.

关键词： Bitcoin

来源：评论

学校读者我要写书评

暂无评论

MediVerse: A Secure and Scalable IoT-MR Framework for Real-Time Health and Performance Monitoring

引用

IEEE Access 2025年 13卷 97511-97528页

作者： Regalado, Pedro H. Han, Tao New Jersey Institute of Technology Electrical and Computer Engineering Department NewarkNJ07102 United States

Recent advancements in wearable and Internet of Things (IoT) technologies have yet to be fully realized in combination with Mixed Reality (MR) for comprehensive real-time health monitoring systems. This paper introduces MediVerse, a secure and scalable IoT-MR framework designed to enhance proactive health management through immersive, context-aware data visualization and real-time feedback. MediVerse employs a three-tiered architecture integrating intelligent sensors, wearable technologies, and MR interfaces to facilitate dynamic prediction and decision-making during critical healthcare scenarios. The framework addresses key challenges in latency, data throughput, interoperability, and user engagement by leveraging edge computing, adaptive compression, and AI-driven analytics. Extensive evaluations demonstrate MediVerse’s superiority, achieving a 70% latency reduction, 51% higher data throughput, and 69% lower error rates compared to conventional systems. Additionally, usability studies highlight significant improvements in user satisfaction and interaction clarity. These findings position MediVerse as a transformative solution for next-generation real-time healthcare monitoring, with adaptability to domains such as sports and education. © 2013 IEEE.

关键词： eHealth

来源：评论

学校读者我要写书评

暂无评论

LRS: Enhancing Adversarial Transferability through Lipschitz Regularized Surrogate 38

LRS: Enhancing Adversarial Transferability through Lipschitz...

引用

38th AAAI Conference on Artificial Intelligence, AAAI 2024

作者： Wu, Tao Luo, Tie Wunsch, Donald C. Department of Computer Science Missouri University of Science and Technology United States Department of Electrical and Computer Engineering Missouri University of Science and Technology United States

ISBN: (纸本)1577358872

The transferability of adversarial examples is of central importance to transfer-based black-box adversarial attacks. Previous works for generating transferable adversarial examples focus on attacking given pretrained surrogate models while the connections between surrogate models and adversarial trasferability have been overlooked. In this paper, we propose Lipschitz Regularized Surrogate (LRS) for transfer-based black-box attacks, a novel approach that transforms surrogate models towards favorable adversarial transferability. Using such transformed surrogate models, any existing transfer-based black-box attack can run without any change, yet achieving much better performance. Specifically, we impose Lipschitz regularization on the loss landscape of surrogate models to enable a smoother and more controlled optimization process for generating more transferable adversarial examples. In addition, this paper also sheds light on the connection between the inner properties of surrogate models and adversarial transferability, where three factors are identified: smaller local Lipschitz constant, smoother loss landscape, and stronger adversarial robustness. We evaluate our proposed LRS approach by attacking state-of-the-art standard deep neural networks and defense models. The results demonstrate significant improvement on the attack success rates and transferability. Our code is available at https://***/TrustAIoT/LRS. Copyright © 2024, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Diversified and compatible web APIs recommendation based on game theory in IoT

引用

Digital Communications and Networks 2024年第4期10卷 1198-1209页

作者： Wenwen Gong Huiping Wu Xiaokang Wang Xuyun Zhang Yawei Wang Yifei Chen Mohammad R.Khosravi College of Information and Electrical Engineering China Agricultural UniversityBeijing100000China Shandong Provincial University Laboratory for Protected Horticulture Weifang University of Science and TechnologyShouguang262700China Department of Computer Science St.Francis Xavier UniversityAntigonishNSCanada Department of Computing Macquarie UniversitySydneyNSW2122Australia Department of Computer Engineering Persian Gulf UniversityBushehr***Iran Department of Electrical and Electronic Engineering Shiraz University of TechnologyShiraz71557-13876Iran

With the ever-increasing popularity of Internet of Things(IoT),massive enterprises are attempting to encapsulate their developed outcomes into various lightweight Web Application Programming Interfaces(APIs)that can be accessible *** this context,finding and writing a list of existing Web APIs that can collectively meet the functional needs of software developers has become a promising approach to economically and easily develop successful mobile ***,the number and diversity of candidate IoT Web APIs places an additional burden on application developers’Web API selection decisions,as it is often a challenging task to simultaneously ensure the diversity and compatibility of the final set of Web APIs *** this challenge and latest successful applications of game theory in IoT,a Diversified and Compatible Web APIs Recommendation approach,namely DivCAR,is put forward in this *** of all,to achieve API diversity,DivCAR employs random walk sampling technique on a pre-built“API-API”correlation graph to generate diverse“API-API”correlation ***,with the diverse“API-API”correlation subgraphs,the compatible Web APIs recommendation problem is modeled as a minimum group Steiner tree search problem.A sorted set of multiple compatible and diverse Web APIs are returned to the application developer by solving the minimum group Steiner tree search *** last,a set of experiments are designed and implemented on a real dataset crawled from *** results validate the effectiveness and efficiency of our proposed DivCAR approach in balancing the Web APIs recommendation diversity and compatibility.

关键词： Internet of things Web APIs recommendation Game theory Diversity and compatibility

来源：评论

学校读者我要写书评

暂无评论

A QoS-based technique for load balancing in green cloud computing using an artificial bee colony algorithm

引用

Journal of Experimental and Theoretical Artificial Intelligence 2025年第2期37卷 307-342页

作者： Tabagchi Milan, Sara Jafari Navimipour, Nima Lohi Bavil, Hamed Yalcin, Senay Department of Computer Engineering Tabriz Branch Islamic Azad University Tabriz Iran Department of Computer Engineering Faculty of Engineering and Natural Sciences Kadir Has University Istanbul Turkey Future Technology Research Center National Yunlin University of Science and Technology Douliou Taiwan Department of Computer Engineering University of Applied Sciences & Technology Tabriz Iran Department of Computer Engineering Nisantasi University Istanbul Turkey

Nowadays, high energy amount is being wasted by computing servers and personal electronic devices, which produce a high amount of carbon dioxide. Thus, it is required to decrease energy usage and pollution. Many applications are utilised by green computing to save energy. Scheduling of tasks acts as an important process to reach the mentioned goals. It is worth stating that the vital characteristic of task scheduling in green clouds is the load balancing of tasks on virtual machines. Efficient load balancing moves tasks from overloaded to underloaded virtual machines to maintain the Quality of Service (QoS). This issue is an NP-complete problem, so this research suggests a new technique based on the behavioural structure of artificial bee behaviour. This method aims to improve QoS while lowering energy usage in green computing. In addition, the honey bees are considered the removed tasks from overloaded virtual machines and a candidate for migrating selected tasks with the lowest priority. The CloudSim testing findings demonstrate that the technique is successful in QoS, makespan, and energy usage compared to other ways. © 2023 Informa UK Limited, trading as Taylor & Francis Group.

关键词： Cloud computing

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：