检索结果-内蒙古大学图书馆

5th International Conference on Artificial Intelligence and Electromechanical Automation, AIEA 2024

作者： Guo, Jinpeng Ding, Jiaxuan Zou, Rongxin Liang, Mingjun School of Computer Science Zhuhai College of Science and Technology Zhuhai China School of Finance and Trade Zhuhai College of Science and Technology Zhuhai China School of Aeronautical Engineering Zhuhai College of Science and Technology Zhuhai China

ISBN: (纸本)9798350366174

In the rapidly developing field of information technology today, the optimization of neural network algorithms has become a common focus of attention in both academia and industry. Especially when dealing with complex data and solving practical problems, optimizing the performance of neural networks is crucial. Therefore, seeking effective optimization algorithms has become the focus of current research. This study is based on the sparrow search algorithm and optimizes the traditional BP neural network to improve its performance in data classification and pattern recognition tasks. Through experimental results analysis, we found that the optimized BP neural network has achieved significant improvements in performance indicators such as accuracy, mean absolute error (MAE), mean square error (MSE), and coefficient of determination (R2), which is significantly better than traditional BP neural networks. Specifically, the optimized model achieved an accuracy of 91.6667%, with a MAE of only 0.012, while the traditional BP neural network was 0.039%;The MSE is 0.014, significantly lower than the 0.042 of traditional BP neural networks;The determination coefficient is 0.93, while the traditional BP neural network is 0.77. These results indicate that the sparrow search algorithm has a significant effect on optimizing BP neural networks and provides important empirical support for improving the performance of neural networks in practical applications. © 2024 IEEE.

关键词： Optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

DoS Attack Detection Based on Deep Factorization Machine in SDN

引用

computer Systems science & engineering 2023年第5期45卷 1727-1742页

作者： Jing Wang Xiangyu Lei Qisheng Jiang Osama Alfarraj Amr Tolba Gwang-jun Kim School of Computer&Communication Engineering Changsha University of Science&TechnologyChangsha410114China Computer Science Department Community CollegeKing Saud UniversityRiyadh11437Saudi Arabia Department of Computer Engineering Chonnam National UniversityGwangju61186Korea

Software-Defined Network(SDN)decouples the control plane of network devices from the data *** alleviating the problems presented in traditional network architectures,it also brings potential security risks,particularly network Denial-of-Service(DoS)*** many research efforts have been devoted to identifying new features for DoS attack detection,detection methods are less accurate in detecting DoS attacks against client hosts due to the high stealth of such *** solve this problem,a new method of DoS attack detection based on Deep Factorization Machine(DeepFM)is proposed in ***,we select the Growth Rate of Max Matched Packets(GRMMP)in SDN as detection ***,the DeepFM algorithm is used to extract features from flow rules and classify them into dense and discrete features to detect DoS *** training,the model can be used to infer whether SDN is under DoS attacks,and a DeepFM-based detection method for DoS attacks against client host is *** results show that our method can effectively detect DoS attacks in *** with the K-Nearest Neighbor(K-NN),Artificial Neural Network(ANN)models,Support Vector Machine(SVM)and Random Forest models,our proposed method outperforms in accuracy,precision and F1 values.

关键词： Software-defined network denial-of-service attacks deep factorization machine GRMMP

来源：评论

学校读者我要写书评

暂无评论

Analytical Model for Performance Measures over Respiratory Track 27

Analytical Model for Performance Measures over Respiratory T...

引用

27th International Symposium on Wireless Personal Multimedia Communications, WPMC 2024

作者： Jaiswal, Raghavendra Singh, S. Pratap Gulhane, Monali Rakesh, Nitin Singh, Rajneesh Kumar Galgotias College of Engineering and Technology Department of Electronics and Communication Engineering Greater Noida India Department of Computer Science and Engineering Pune India Sharda University Sharda School of Engineering and Technology Greater Noida India

ISBN: (纸本)9798350392319

Diffusive Molecular Communication (DMC) as a framework for modeling bacterial propagation within biological environments. DMC, a subset of Molecular Communication (MC), allows us to study how bacteria transmit signals inside the human body. The effectiveness of the model depends on the chosen deployment geometry. We focus on a cylindrical biological deployment inspired by the human respiratory tract. This model considers the degradation and flow of information molecules diffusing through the fluid medium. Furthermore, proposed statistical models for bacterial concentration f(y) Further, analyse the key performance metrics within this CBD model and present statistical models for bacterial propagation are binding probability (Pb), non-binding probability (Pnb), evade probability (Pe(y)), and the probability of binding rate (Pb(y)). Major contribution of numerical analysis to find total number of receptor bacteria at y segment N(y), and Bacterial distribution V(y). Further, the numerical results demonstrate excellent agreement with the theoretical foundation. © 2024 IEEE.

关键词： Bacterial distribution binding probability cylindrical biological deployment Diffusive Molecular Communication

来源：评论

学校读者我要写书评

暂无评论

Private Data Aggregation Enabling Verifiable Multisubset Dynamic Billing in Smart Grids 23

Private Data Aggregation Enabling Verifiable Multisubset Dyn...

引用

23rd IEEE International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2024

作者： Yang, Qian Wang, Chen Shen, Jian Li, Yi Liu, Dengzhi Zhejiang Sci-Tech University School of Computer Science and Technology HangZhou310018 China Swinburne University of Technology School of Software and Electrical Engineering HawthornVIC Australia Jiangsu Ocean University School of Computer Engineering Lianyungang China

Efficient power management in smart grids relies on regularly obtaining the electricity usage of each user. However, the aggregation of the electricity usage data may expose user privacy. At present, most existing solutions aggregate the electricity data of the entire user set, which cannot meet the fine-grained requirements of the control center. Therefore, this paper proposes a verifiable privacy-preserving multisubset dynamic billing data aggregation scheme. Firstly, we divide the electricity data into k consecutive subsets and set dynamic pricing rules, such as pricing within a certain range as q1 and exceeding it as q2. Then, the smart meter encrypts the data using the Paillier cryptographic system with the corresponding subset of parameters and uploads them to the aggregation equipment. After the aggregation equipment completes the users' data aggregation, the ciphertext C is sent to the control center. This process enables the control center to check data integrity and obtain the total number of people, electricity consumption, and costs for different ranges of electricity in a period of time at once, without the need to obtain data for individual users. Analysis and experiments show that this scheme can resist attacks from powerful adversaries on the transmission channel and has practicality and effectiveness. © 2024 IEEE.

关键词： Sensitive data

来源：评论

学校读者我要写书评

暂无评论

Research on Pulmonary Nodule Detection Method Based on YOLO Algorithm 5

Research on Pulmonary Nodule Detection Method Based on YOLO ...

引用

5th International Seminar on Artificial Intelligence, Networking and Information technology, AINIT 2024

作者： Ma, Zewei Wu, Feier School of Cyber Science and Engineering Nanjing University of Science and Technology Nanjing China School of Artificial Intelligence and Computer Science Jiangnan University Wuxi China

ISBN: (纸本)9798350385557

This paper presents a method for pulmonary nodule detection in CT images based on an improved YOLOv8 framework. Against the backdrop of current deep learning technologies, automated detection and segmentation of medical images are crucial for enhancing diagnostic speed and accuracy. We combine the YOLOv8 object detection framework with Receptive Fields Block (RFB) modules to enhance feature extraction capabilities. Additionally, we introduce Inner-IoU as a loss function to optimize the detector's localization accuracy. Experimental results demonstrate significant performance improvement in pulmonary nodule detection tasks on CT images, thereby providing substantial support for the automation and intelligence of medical image processing. © 2024 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Integrated deep learning and Bayesian optimization approach for enhanced prediction of high-performance concrete strength

引用

Asian Journal of Civil engineering 2025年第6期26卷 2371-2390页

作者： Tipu, Rupesh Kumar Goyal, Archna Singh, Digvijay Kumar, Ayyala Kishore Ajay Department of Civil Engineering School of Engineering & Technology K. R. Mangalam University Sohna Haryana Gurugram 122103 India Department of Computer Science & Engineering School of Engineering & Technology K. R. Mangalam University Sohna Haryana Gurugram 122103 India Department of Electrical Engineering School of Engineering & Technology K. R. Mangalam University Sohna Haryana Gurugram 122103 India

Prediction of high performance concrete (HPC) compressive strength is very important for determination of mix design optimization and structural reliability. An integrated deep learning framework, together with a Bayesian optimization for improving accuracy in prediction, is what this study presents. Finally, the optimized DNN yielded a coefficient of determination (R2) of 0.932 and the lowest RMSE (4.104 MPa) among all the proposed regression models, and was superior to the traditional regression models, including linear regression (RMSE = 10.089 MPa) and kernel ridge regression (RMSE = 10.095 MPa). Hyperparameters were fine tuned using a Bayesian optimization reducing the RMSE from 10.985 MPa (unoptimized DNN) down to 4.104 MPa. SHAP-based feature importance analysis revealed that age and cement content were the most influential variables, reinforcing domain knowledge about cement hydration and strength development. In addition, a graphical user interface (GUI) was built to make practical implementation possible and thus allow real time compressive strength prediction using material proportions. In summary, it shows that a deep learning combined with hyperparameter optimization is able to greatly improve the predictive reliability and efficiency of the HPC strength, which facilitates the sustainable construction practices and efficient material utilization. © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2025.

关键词： Bayesian optimization Compressive strength prediction Deep neural networks High-performance concrete SHAP analysis

来源：评论

学校读者我要写书评

暂无评论

UAV Path Planning with an Adaptive Hybrid PSO 2

UAV Path Planning with an Adaptive Hybrid PSO

引用

2nd International Conference on Information and Communication technology for Sustainable Development, ICICT4SD 2023

作者： Nayeem, Golam Moktader Fan, Mingyu Daiyan, Golam Moktader Fahad, Khaled Saifullah School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu611731 China Bangladesh Army International University of Science and Technology Department of Computer Science and Engineering Cumilla Bangladesh

ISBN: (纸本)9798350358667

Optimized Path planning plays a significant role in enabling autonomous operation of Unmanned Aerial Vehicles (UAVs). Finding the shortest path is one of many NP-hard problems for which metaheuristic algorithms have shown promising results. Particle swarm optimization (PSO) is a frequently used metaheuristic approach for robotic path planning. PSO, however, has significant flaws like poor exploration capacity and a lack of diversity in the particles. PSO may become vulnerable to local minima due to either of these constraints. Inertia weight formulation and hybridization with other metaheuristic algorithms are two popular strategies for overcoming PSO's drawbacks. This study presents the aw-GPSO adaptive hybrid PSO algorithm, which is applied for offline 3D path planning of UAVs. To enhance the exploration capability of the PSO algorithm and improve particle diversity, we incorporate the Grey Wolf Optimization (GWO) algorithm and introduce a time-varying adaptive inertia weight parameter. To showcase the enhancements achieved, we conducted simulations of our proposed approach and conducted a comparative analysis with recent variants of PSO utilized in UAV path planning. © 2023 IEEE.

关键词： Motion planning

来源：评论

学校读者我要写书评

暂无评论

Real-time Obesity and Heart Disease Monitoring with IoT and Deep Learning

Real-time Obesity and Heart Disease Monitoring with IoT and ...

引用

2024 International Conference on Smart Devices, ICSD 2024

作者： Singh, Jagendra Govindaraju, P. Kumar, Vinish Yamsani, Nagendar Umamaheswari, J. Munagala, Madhu Bennett University School of Computer Science Engineering & Technology Greater Noida India K S R Institute for Engineering and Technology Department of Electronics and Communication Engineering Tiruchengode India Raj Kumar Goel Institute of Technology Department of Computer Science Engineering Ghaziabad India SR University School of Computer Science and Artificial Intelligence Warangal India Brainware University Department of Computer Science Kolkata India Koneru Lakshmaiah Education Foundation Department of of English Vaddeswaram India

ISBN: (纸本)9798350394412

The accelerating spread of junk food consumption is progressively putting the younger generations at risk of obesity. The concept of health prioritization is deeply rooted in the development of nurturing and general welfare in society. The job has further become more complex by the glaring statistics of people across the globe who are fighting obesity. The World Health Organization (WHO) puts a staggering estimate that approximately 1 billion people across the globe suffer from obesity-related issues, a statistic heavy in terms of impact primarily because obesity greatly affects one's health but also holds fatal risks. Diagnosing the early stages of obesity can just be simple and prevent one from being obese;however, several obstacles stand in the way of its early diagnosis. Symptoms or signs that one is in the trajectory of suffering from obesity generally go unnoticed by the individual, and the availability of specialists to diagnose such risks persistently on time is not always possible. In this regard, the current paper shall showcase an application hitherto unknown using cloud-based diagnosis, in collaboration with machine learning through Secure Access Internet of Things to predict obesity-related risks without having to depend on glaring physical parameter indicators - in this regard, redefining how such information could be sourced even without the need of regularly visiting health professional people. To ensure the best possible health predictability from this machine learning, we gathered and plotted the datasets consisting of obese and overweight people. Other than machine learning, we use a Secure Access IoT sensor network. This allows for real-time monitoring and diagnosis, making early detection achievable. Consequently, this system paves the way for mitigating and substantially reducing the complications associated with obesity. © 2024 IEEE.

关键词： Nutrition

来源：评论

学校读者我要写书评

暂无评论

Artificial Intelligence-based Solution for the Prediction of Power Consumption in Electronics and Software Applications

引用

IETE Journal of Research 2024年第1期70卷 356-371页

作者： C, Savitha Khampariya, Prabodh Singh, Kamred Udham Kumar, Ankit Singh, Teekam Swarup, Chetan Department of Electronics and Communication Engineering Sri Satya Sai University of Technology & Medical Sciences Sehore India Department of Computer Science and Information Engineering National Cheng Kung University Tainan Taiwan Computer Engineering and Applications GLA University Mathura281 406 India School of Computer Science University of Petroleum and Energy Studies Dehradun India Basic Science College of Science and Theoretical Studies Saudi Electronic University Riyadh-Male Campus Riyadh11673 Saudi Arabia

As new computing paradigms such as mobile grids and clouds become more commonplace, mobile devices are becoming increasingly attractive to scientists and HPC users who need high-performance computing capabilities. There are still many challenges to design software that uses mobile devices, such as their restricted capabilities compared to traditional devices such as PCs and servers. It’s even more critical to remember that batteries are the primary power source for mobile devices. Thus people delete programmes that severely reduce their battery life. The upshot is that it is common for developers to have no idea how much power specific hardware and software components use. So even if they have been trained in software development, they still require specific standards and skills to design energy-efficient apps. This Paper investigates ways to minimize mobile device energy usage by restructuring source code. However, we did not overlook the prospect of applying our findings to other domains, such as recurring demanding computational cores in scientific applications. Specifically, this research contributes to the field of Green Computing and mobile software development. © 2024 IETE.

关键词： Energy efficiency

来源：评论

学校读者我要写书评

暂无评论

Leveraging Machine Learning for Enhanced SensorPowered Green IoT with Federated Learning 15

Leveraging Machine Learning for Enhanced SensorPowered Green...

引用

15th International Conference on Computing Communication and Networking Technologies, ICCCNT 2024

作者： Devrani, Rekha Premkumar, R. Kumar, Shyam Vivekananda Global University Department of Computer Science & Engineering Jaipur India School of Engineering and Technology Department of Electronics and Communication Engineering Karnataka Bangalore India Arka Jain University Department of Computer Science & It Jharkhand Jamshedpur India

ISBN: (纸本)9798350370249

Federated getting to know (FL) is a gadget studying generation which allows a user to train a single model across more than one dispensed system, with none of the person systems having to transfer the statistics sets, nor the version, to a vital device for education. It allows disbursed schooling of an unmarried model throughout more than one disbursed net of things (IoT) gadgets or users. It makes it feasible to improve the accuracy and scalability of our Sensor-Powered inexperienced IoT applications. FL can consequently be instrumental in taking inexperienced IoT answers towards the character sources of environmental records, like air firstclass sensors, as opposed to amassing facts *** permits us to leverage powerful gadget learning models to enhance the accuracy and scalability of our Sensor-Powered green IoT systems. By using federated models throughout education, we will grow the accuracy of our predictions and decrease the amount of effort and time we have to acquire and store sensor facts centrally. Moreover, FL lets us improve the version overall performance of green IoT packages by dispensing schooling and predictions to a disbursed community of sensor-based totally IoT devices. We also can, in addition, reduce the reliance on central systems with the aid of transferring the middle of our Sensor-Powered inexperienced IoT packages from the cloud to allotted nodes. © 2024 IEEE.

关键词： Generation Predictions Scalability Sensor Statistics Throughout Unmarried

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：