检索结果-内蒙古大学图书馆

2024 Innovations in Intelligent Systems and Applications Conference, ASYU 2024

作者： Irican, Beyza Belen Sivri, Müslüm Kokach, Vael Kocaçınar, Büşra Akbulut, Fatma Patlar Department of Computer Engineering Istanbul Kültür University Istanbul Turkey Department of Software Engineering Istanbul Kültür University Istanbul Turkey

ISBN: (纸本)9798350379433

Chatbots are becoming more and more popular as efficient tools in information retrieval and customer service procedures, which raises satisfaction by offering quick and efficient answers. This work presents a chatbot system designed to handle lengthy and intricate documents and to accurately and semantically respond to user queries. Big language models and vector-based search methods were combined in the Retrieval Augmented Generation (RAG) approach of system design. For precise and meaningful answers to user queries, the system dynamically retrieves data and interfaces with OpenAI API. Regarding chatbot system response quality and user satisfaction, this work can be regarded as a significant step. © 2024 IEEE.

关键词： Customer satisfaction

来源：评论

学校读者我要写书评

暂无评论

Markov Chain-based Data Privacy Prioritizing for Task Scheduling in Digital Twin Edge Networks 6

Markov Chain-based Data Privacy Prioritizing for Task Schedu...

引用

6th International Conference on Blockchain Computing and Applications, BCCA 2024

作者： Bozkaya-Aras, Elif Erel-Özçevik, Müge Department of Computer Engineering National Defence University Turkish Naval Academy Istanbul Turkey Department of Software Engineering Manisa Celal Bayar University Manisa Turkey

ISBN: (纸本)9798350351538

Edge computing plays a significant role in shaping the future of Internet of Things (IoT) networks by processing high volumes of data at the network edge and meeting the desired latency demand. Although edge computing is regarded as a promising technology for these features as well as its decentralization and supportive functionalities, there is a need to develop systems to protect data security and privacy since users can face the risk of data leakage when they offload data to the edge/cloud server. In order to solve this problem, in this paper, we propose a three-layer digital twin edge network architecture for task scheduling based on edge-cloud computing. The digital twin has been used to map the IoT network to the virtual space, monitor physical assets in real-time, and perform comprehensive network analysis. On this basis, a task scheduling model that prioritizes data privacy has been introduced to prevent users’ privacy-sensitive information from being collected by attackers. The proposed model can both take full advantage of edge and cloud computing and protect data privacy. We classify computation tasks based on data privacy sensitivity and model them using the M/M/1 Markov model in local, edge server, and cloud computations. Then, we propose a data privacy prioritizing task scheduling algorithm combining digital twin and edge-cloud collaboration. This genetic optimization based algorithm significantly increases data privacy by 40% compared to the greedy approach, while keeping the time under an acceptable level. In this way, each task is distributed between the local, edge server, and cloud server. Thanks to the intelligence behavior and repetitive growth of the proposed algorithm, the optimal task scheduling is performed in a digital twin system. Our analysis reveals that data leakage attacks can be significantly decreased with the proposed task scheduling algorithm by minimizing task processing time. © 2024 IEEE.

关键词： Differential privacy

来源：评论

学校读者我要写书评

暂无评论

Hyperparameter Optimization in Deep Learning-Based Object Detection of Branching and Endpoints on 2D Brain Vessel Images

Hyperparameter Optimization in Deep Learning-Based Object De...

引用

2024 Innovations in Intelligent Systems and Applications Conference, ASYU 2024

作者： Kaya, Samet Kiraz, Berna Çamurcu, Ali Yılmaz Department of Software Engineering Fatih Sultan Mehmet Vakif University Istanbul Turkey Department of Computer Engineering Fatih Sultan Mehmet Vakif University Istanbul Turkey

ISBN: (纸本)9798350379433

This work presents a deep learning-based object detection technique for identifying branches and endpoints in two-dimensional brain vessel images alongside its hyperparameter optimization. Although traditional image processing methods are feasible and successful, their algorithm complexity increases exponentially with the size of the image and the filters to be applied. In contrast, our deep learning approach has shown significant improvements in accuracy and efficiency, independent of image size and number of filters. We preprocess our dataset of raw mouse brain slices from laboratory environments to remove noise from images and then extract a binary vein network using image processing methods. Finally, we labeled vessel branching and endpoints in 5x5 pixel bounding boxes. All labeled objects on images were converted to COCO format for training and testing to ensure compatibility with deep learning algorithms. Our research focused on using the Faster R-CNN method in the Detectron2 framework, which has been successful in our previous work. Evaluation using the intersection over union (IoU) metric underscores the robustness of our approach, and we achieved a success rate of over 90%. We employed Optuna for hyperparameter optimization to further enhance our model, focusing on three key hyperparameters: base learning rate, maximum iterations, and batch size per image. We systematically refined these hyperparameters by running 50 training and test runs separately, significantly improving model performance to over 98%. Our findings highlight the transformative potential of deep learning in neuroimaging analysis and promise significant advances in diagnostic and research applications in neurovascular studies. © 2024 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

A Novel Deep Federated Learning-Based and Profit-Driven Service Caching Method 19th

A Novel Deep Federated Learning-Based and Profit-Driven Ser...

引用

19th EAI International Conference on Collaborative Computing: Networking, Applications and Worksharing, CollaborateCom 2023

作者： Ouyang, Zhaobin Xia, Yunni Peng, Qinglan Li, Yin Chen, Peng Wang, Xu College of Computer Science Chongqing University Chongqing China School of Artificial Intelligence Henan University Zhengzhou China Guangzhou Institute of Software Application Technology Guangzhou China School of Computer and Software Engineering Xihua University Chengdu China College of Mechanical and Vehicle Engineering Chongqing University Chongqing China

ISBN: (纸本)9783031545306

Service caching is an emerging solution to addressing massive service request in a distributed environment for supporting rapidly growing services and applications. With the explosive increases in global mobile data traffic, service caching over the edge computing architecture, Mobile edge computing (MEC), emerges for alleviating traffic congestion as well as for optimizing the efficiency of task processing. In this manuscript, we propose a novel profit-driven service caching method based on a federated learning model for service prediction and a deep reinforcement learning mode for yielding caching decisions (FPDRD) in an edge environment. The proposed method is temporal service popularity and user preference-aware. It aims to ensure quality of service (QoS) of delivery of cached service while maximizing the profits of network service providers. Experimental results clearly demonstrate that the FPDRD method outperforms traditional methods in multiple aspects. © ICST Institute for computer Sciences, Social Informatics and Telecommunications engineering 2024.

关键词： Traffic congestion

来源：评论

学校读者我要写书评

暂无评论

Performance Analysis of Cryptographic Algorithms in Cloud Computing Systems 7

Performance Analysis of Cryptographic Algorithms in Cloud Co...

引用

7th International Conference on Inventive Computation Technologies, ICICT 2024

作者： Mohammed, Abdul Sami Alfudail, Mohammed Alkhulaiwi, Abdullah Bashar, Abul Prince Mohammad Bin Fahd University Computer Engineering Khobar Saudi Arabia Prince Mohammad Bin Fahd University Software Engineering Khobar Saudi Arabia

ISBN: (纸本)9798350359299

Cloud computing has rapidly evolved into a fundamental paradigm for data storage and processing in contemporary IT enterprises. This growth has heightened concerns regarding the security of sensitive data in cloud environments. In response to this, this research study investigates and compares five prominent cryptographic algorithms, including DES, 3-DES, AES, RSA, and SHA-256, in terms of their performance and efficiency. Leveraging the PyCryptodome library, we conducted a comprehensive analysis of these algorithms and their suitability for cloud systems. The results reveal important insights into their practical applicability, with implications for data security in cloud computing. Our findings underscore the superior performance of AES, positioning it as a robust choice for various cloud applications. © 2024 IEEE.

关键词： Cloud security

来源：评论

学校读者我要写书评

暂无评论

Impact of Artificial Intelligence on software engineering Phases and Activities (2013–2024): A Quantitative Analysis Using Zero- Truncated Poisson Model

引用

IEEE Access 2025年 13卷 95535-95547页

作者： Durrani, Usman Khan Akpinar, Mustafa Bektas, Hakan Saleh, Mohammed Higher Colleges of Technology Faculty of Computer and Information Sciences Abu Dhabi United Arab Emirates Sakarya University Software Engineering Department Sakarya54050 Turkey Istanbul University Department of Econometrics Istanbul34452 Turkey

This paper presents the results of a quantitative analysis derived from data collected in our earlier systematic literature review, focusing on integrating Artificial Intelligence (AI) techniques across various phases of software engineering (SE). Employing a Zero-Truncated Poisson (ZTP) model, we analyzed 120 selected research papers to assess the impact of AI methodologies on SE phases. The findings indicate significant improvements attributable to AI in different phases of SE. Specifically, AI techniques have been found to enhance the accuracy of the planning, requirement engineering, and development phases and improve the efficiency of the requirement engineering and design phases. Notably, AI’s integration has markedly improved the accuracy and efficiency of the requirement engineering phase, underscoring AI’s critical role in advancing SE practices. This study contributes to a deeper understanding of the quantitative impact of AI on SE, providing valuable insights for practitioners and researchers in understanding AI technologies to enhance SE phases. © 2013 IEEE.

关键词： software design

来源：评论

学校读者我要写书评

暂无评论

Dynamic Multimedia Encryption Using a Parallel File System Based on Multi-Core Processors 4

Dynamic Multimedia Encryption Using a Parallel File System B...

引用

4th IEEE International Conference on Data engineering and Communication Systems, ICDECS 2024

作者： Simi Margarat, G. Malathi, S. Pasumpon Pandian, A. Anakath, A.S. Kannadasan, R. New Prince Shri Bhavani College of Engineering and Technology Department of Computer Science and Engineering Tamil Nadu Chennai India Grt Institute of Engineering and Technology Department of Computer Science and Engineering Tamil Nadu Tiruttani India Care College of Engineering Department of Computer Science and Engineering Trichy India Saveetha Institute of Medical and Technical Sciences Saveetha School of Engineering Department of Computer Science and Engineering Tamil Nadu Chennai India School of Computer Science and Engineering Department of Software Systems Tamil Nadu Vellore India

ISBN: (纸本)9798350393354

A new parallel file system and multi-core processor-based dynamic multimedia encryption method is presented in this study. Multimedia encryption efficiency and security were the main goals, addressing massive data sets and high-speed processing. We created a dynamic encryption algorithm for multimedia content, supported by a strong parallel file system and tuned for multi-core processing. Multiple tests showed significant improvements in encryption and decryption times, retaining short durations for real-time applications. After encryption, the system maintained multimedia content quality and accuracy with data integrity rates nearing 100%. Balanced CPU utilization showed efficient resource use and processor load management. The system's excellent throughput efficiency showed its capacity to process enormous amounts of data. These results indicate that parallel file systems and multi-core CPUs may solve multimedia encryption problems. Adaptable and scalable, the system supports many multimedia formats and sizes. This paper proposes a multimedia encryption system that improves performance data security and integrity, paving the way for safe multimedia processing advances. © 2024 IEEE.

关键词： Cryptography

来源：评论

学校读者我要写书评

暂无评论

A Trust-based Framework for the Detection and Avoidance of Blackhole Nodes in Opportunistic Networks 9

A Trust-based Framework for the Detection and Avoidance of B...

引用

9th International Conference on Signal Processing and Communication, ICSC 2023

作者： Kundu, Protyasha Saha, Swarnadeep Kundu, Anindita Barai, Smritikona Department of Computer Science and Engineering Heritage Institute of Technology Kolkata India Department of Software Systems School of Computer Science and Engineering Vellore Institute of Technology Vellore India

ISBN: (纸本)9798350383201

An Opportunistic Network (OppNet), as opposed to a ubiquitous centralized network, relies on sporadic and opportunistic encounters between nodes to facilitate communication. The uncertainty about the node's nature presents challenges when nodes with malicious intents are present in the network. One category of malicious nodes includes blackhole nodes, which receive messages but drop them intentionally. In this paper, we introduce a novel trust function designed to detect blackhole nodes within the network. Our function has been widely tested and simulated to validate and increase the efficacy of transmission of messages in the network. We have performed a comparative analysis with various protocols (PROPHET, Epidemic and Spray And Wait) on several parameters such as Delivery Ratio, Overhead Ratio and Number of Messages dropped. The results demonstrate that our trust function is able to correctly detect and avoid blackhole nodes. We observed significant improvement in delivery ratio in Epidemic Routing (47%), Prophet (41%), and Spray And Wait Routing (17%) protocols when adjusting blackhole nodes while keeping the total network nodes constant. © 2023 IEEE.

关键词：

来源：评论

学校读者我要写书评

暂无评论

M-GhostNet: A Lightweight CNN Model Combined with Coordinate Attention Mechanism for Identifying Pests and Diseases 12

M-GhostNet: A Lightweight CNN Model Combined with Coordinate...

引用

12th International Conference on Computing and Pattern Recognition, ICCPR 2023

作者： Ma, Chi Xiao, Yu Hu, Hui Li, Jingyan Wu, Jie The School of Computer Science and Engineering Huizhou University Huizhou516007 China The School of Computer Science and Software Engineering University of Science and Technology Liaoning Anshan114051 China

ISBN: (纸本)9798400707988

Plant pests and diseases have always been an important factor affecting agricultural production, and how to automatically identify them is one of the current research hotspots. Although traditional deep learning models can achieve good results, they are difficult to deploy into small devices due to their large number of parameters. In this paper, we design a lightweight convolutional neural network M-GhostNet based on the coordinate attention mechanism, which improves the bottleneck structure in the original GhostNet and and effectively extracts the spatial information of the image. We compare and ablate the M-GhostNet model on the IP102 dataset and the Embrapa dataset. Compared with the existing lightweight neural networks, M-GhostNet achieves the best results of 67.15%, 65.04%, and 71.41% on the IP102 dataset, respectively, and 92.67%, 95.65%, 93.74%, and 96.52% on the Embrapa dataset in terms of Mrec, Mpre, MF1, and Acc. The final results show that M-GhostNet can achieve good performance in both pest and disease identification. © 2023 ACM.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

RoSe: Rotation-Invariant Sequence-Aware Consensus for Robust Correspondence Pruning 24

RoSe: Rotation-Invariant Sequence-Aware Consensus for Robust...

引用

32nd ACM International Conference on Multimedia, MM 2024

作者： Liu, Yizhang Zhou, Weiwei Li, Yanping Zhao, Shengjie School of Software Engineering Tongji University Shanghai China College of Computer and Data Science Fuzhou University Fuzhou China Department of Computer Science and Technology Tongji University Shanghai China Engineering Research Center of Key Software Technologies for Smart City Perception and Planning Ministry of Education Shanghai China

ISBN: (纸本)9798400706868

Correspondence pruning has recently drawn considerable attention as a crucial step in image matching. Existing methods typically achieve this by constructing neighborhoods for each feature point and imposing neighborhood consistency. However, the nearest-neighbor matching strategy often results in numerous many-to-one correspondences, thereby reducing the reliability of neighborhood information. Furthermore, the smoothness constraint fails in cases of large-scale rotations, leading to misjudgments. To address the above issues, this paper proposes a novel robust correspondence pruning method termed RoSe, which is based on rotation-invariant sequence-aware consensus. We formulate the correspondence pruning problem as a mathematical optimization problem and derive a closed-form solution. Specifically, we devise a rectified local neighborhood construction strategy that effectively enlarges the distribution between inliers and outliers. Meanwhile, to accommodate large-scale rotation, we propose a relative sequence-aware consistency as an alternative to existing smoothness constraints, which can better characterize the topological structure of inliers. Experimental results on image matching and registration tasks demonstrate the effectiveness of our method. Robustness analysis involving diverse feature descriptors and varying rotation degrees further showcases the efficacy of our method. © 2024 ACM.

关键词： correspondence pruning

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：