检索结果-内蒙古大学图书馆

International Conference on Open Source Systems and Technologies (ICOSST)

作者： Soomaiya Hamid Narmeen Zakaria Bawany Center for Computing Research Department of Computer Science and Software Engineering Jinnah University for Women Karachi Pakistan

ISBN: (数字)9798331508692

ISBN: (纸本)9798331508708

The integration of IoT devices in smart cities enhances urban infrastructure, services, and governance but also introduces significant cybersecurity challenges. Traditional centralized Intrusion Detection Systems (IDS) face several issues, including data privacy concerns and high-power consumption due to centralized data processing. These challenges increase the risks of unauthorized access, data breaches, and privacy violations, undermining user trust and compliance with privacy regulations. Additionally, the centralization of data and processing leads to higher power consumption, making these systems less sustainable for widespread deployment in smart cities. This research addresses these issues by proposing a Federated Learning (FL)based intrusion detection framework for smart cities. FL enables collaborative and privacy-preserving model training across distributed IoT devices, mitigating the need to share sensitive data centrally. By aggregating local model updates, FL ensures data privacy and distributes the computational workload, significantly reducing power consumption compared to traditional centralized systems. The proposed model leverages advanced AI techniques and is trained using the IoTID20 dataset. The Flower framework, utilizing the FedAvg algorithm, facilitates the federated learning process. Our experimental results demonstrate that the global model achieves 98% accuracy, with individual clients achieving accuracies of around 85% to 98%. This approach provides continuous learning mechanisms, anomaly detection, and ensemble learning capabilities, enhancing the resilience of federated intrusion detection systems against emerging threats and adversarial attacks. This research systematically investigates the application of federated learning for intrusion detection in smart city networks, addressing key challenges and advancing the state-of-the-art in decentralized cybersecurity solutions. The proposed framework offers a robust, scalable, and privacyco

关键词： Data privacy Power demand Accuracy Smart cities Federated learning Computational modeling Intrusion detection Data models Internet of Things Computer security

来源：评论

学校读者我要写书评

暂无评论

Toward Privacy-Preserving Spatial Crowdsourcing: From Offline to Online

引用

IEEE Network 2024年第3期39卷 326-332页

作者： Hengzhi Wang Junjie Mai Lei Zhang Linfeng Shen Laizhong Cui Jiangchuan Liu College of Computer Science and Software Engineering Shenzhen University Shenzhen China School of Computing Science Simon Fraser University Burnaby BC Canada

Recent years have seen Spatial Crowdsourcing (SC) becoming a promising paradigm of collecting large amounts of location-aware data for various network applications. Its strengths lie in leveraging the power of the crowd to gather data effectively at a low cost. However, during the interaction process between SC participants (e.g., workers and task requestors) and the SC platform, participants’ sensitive information is inevitably exposed, raising severe privacy concerns. Existing studies addressing these privacy concerns have mostly focused on offline SC settings, where the participants’ information is known and accessible in advance. Yet real-world SC participants can be highly dynamic with uncertain behaviors; hence, in practice, their information is largely unknown a priori. In this article, we closely examine the privacy challenges as well as the trade-off between privacy and efficiency in online SC settings. We develop a comprehensive online privacy-preserving SC framework and analyze the critical implementation issues.

关键词： Privacy Costs Crowdsourcing Real-time systems Protection Vehicle dynamics Data privacy

来源：评论

学校读者我要写书评

暂无评论

Huffman Tree Compression and Lempel-Ziv Coding Using Java

Huffman Tree Compression and Lempel-Ziv Coding Using Java

引用

Decision Aid Sciences and Application (DASA), International Conference on

作者： Hilal H. Nuha Hassan Rizky Putra Sailellah Ahmed Abo Absa School of Computing Telkom University Bandung Indonesia Software Engineering University of Palestine Gaza Strip Palestine

ISBN: (数字)9798350369106

ISBN: (纸本)9798350369113

Lossless data compression is necessary to reduce transmission costs while maintaining data integrity. This paper describes Kompressor as a Huffman and Lempel Ziv (LZ) compression algorithm project using Java. All major Java class implementations are discussed in detail. The algorithm is evaluated for three different files to measure performance such as compression ratio, code rate, and efficiency. The program is shown to be capable of performing compression and decompression as proposed in the original algorithm for different parameter values. Huffman encoding, in particular, is capable of shrinking files to less than 70% of their original file size. It is hoped that this project will be useful for academic and scientific purposes, especially for courses based on data compression.

关键词： Degradation Java Computer languages Costs Data integrity Symbols Data compression Propagation losses Particle measurements Encoding

来源：评论

学校读者我要写书评

暂无评论

PERIODICITY DECOUPLING FRAMEWORK FOR LONG-TERM SERIES FORECASTING 12

PERIODICITY DECOUPLING FRAMEWORK FOR LONG-TERM SERIES FORECA...

引用

12th International Conference on Learning Representations, ICLR 2024

作者： Dai, Tao Wu, Beiliang Liu, Peiyuan Li, Naiqi Bao, Jigang Jiang, Yong Xia, Shu-Tao College of Computer Science and Software Engineering Shenzhen University China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University China Tsinghua Shenzhen International Graduate School Tsinghua University Shenzhen China WeBank Institute of Financial Technology Shenzhen University China

Convolutional neural network (CNN)-based and Transformer-based methods have recently made significant strides in time series forecasting, which excel at modeling local temporal variations or capturing long-term dependencies. However, real-world time series usually contain intricate temporal patterns, thus making it challenging for existing methods that mainly focus on temporal variations modeling from the 1D time series directly. Based on the intrinsic periodicity of time series, we propose a novel Periodicity Decoupling Framework (PDF) to capture 2D temporal variations of decoupled series for long-term series forecasting. Our PDF mainly consists of three components: multi-periodic decoupling block (MDB), dual variations modeling block (DVMB), and variations aggregation block (VAB). Unlike the previous methods that model 1D temporal variations, our PDF mainly models 2D temporal variations, decoupled from 1D time series by MDB. After that, DVMB attempts to further capture short-term and long-term variations, followed by VAB to make final predictions. Extensive experimental results across seven real-world long-term time series datasets demonstrate the superiority of our method over other state-of-the-art methods, in terms of both forecasting performance and computational efficiency. Code is available at https://***/Hank0626/PDF. © 2024 12th International Conference on Learning Representations, ICLR 2024. All rights reserved.

关键词： Time series

来源：评论

学校读者我要写书评

暂无评论

Adaptive Activation Steering: A Tuning-Free LLM Truthfulness Improvement Method for Diverse Hallucinations Categories 25

Adaptive Activation Steering: A Tuning-Free LLM Truthfulness...

引用

34th ACM Web Conference, WWW 2025

作者： Wang, Tianlong Jiao, Xianfeng Zhu, Yinghao Chen, Zhongzhi He, Yifan Chu, Xu Gao, Junyi Wang, Yasha Ma, Liantao School of Software and Microelectronics Peking University Beijing China Key Laboratory of High Confidence Software Technologies Ministry of Education Beijing China National Engineering Research Center for Software Engineering Peking University Beijing China Beihang University Beijing China Center on Frontiers of Computing Studies Peking University Beijing China Centre for Medical Informatics University of Edinburgh Edinburgh United Kingdom Health Data Research UK London United Kingdom Key Laboratory of High Confidence Software Technologies Ministry of Education National Engineering Research Center for Software Engineering Peking University Beijing China

ISBN: (纸本)9798400712746

Recent studies have indicated that Large Language Models (LLMs) harbor an inherent understanding of truthfulness, yet often fail to consistently express it and generate false statements. This gap between "knowing" and "telling" poses a challenge for ensuring the truthfulness of generated content. Inspired by recent work on the practice of encoding human-interpretable concepts linearly within large language models, we treat truthfulness as a specially linearly encoded concept within LLMs, and introduce Adaptive Activation Steering (ACT), a tuning-free method that adaptively shifts LLM’s activations in the "truthful" direction during inference. ACT addresses diverse categories of hallucinations by utilizing diverse truthfulness-related steering vectors and adjusting the steering intensity adaptively. Applied as an add-on across various models, ACT significantly improves truthfulness in LLaMA (↑ 142%), LLaMA2 (↑ 24%), Alpaca (↑ 36%), Vicuna (↑ 28%), LLaMA2-Chat (↑ 19%), and LLaMA3(↑ 34%). Furthermore, we verify ACT’s scalability across larger models (13B, 33B, 65B), underscoring the adaptability of ACT to large-scale language models. © 2025 Copyright held by the owner/author(s).

关键词： hallucination

来源：评论

学校读者我要写书评

暂无评论

Building Personalised XAI Experiences Through iSee: a Case-Based Reasoning-Driven Platform 2

Building Personalised XAI Experiences Through iSee: a Case-B...

引用

Joint of the 2nd World Conference on eXplainable Artificial Intelligence Late-Breaking Work, Demos and Doctoral Consortium, xAI-2024:LB/D/DC

作者： Caro-Martínez, Marta Liret, Anne Díaz-Agudo, Belén Recio-García, Juan A. Darias, Jesús Wiratunga, Nirmalie Wijekoon, Anjana Martin, Kyle Nkisi-Orji, Ikechukwu Corsar, David Palihawadana, Chamath Pirie, Craig Bridge, Derek Pradeep, Preeja Fleisch, Bruno Department of Software Engineering and Artificial Intelligence Universidad Complutense de Madrid Spain School of Computing Robert Gordon University Aberdeen United Kingdom School of Computer Science & IT University College Cork Ireland British Telecommunications United Kingdom

Nowadays, eXplainable Artificial Intelligence (XAI) is well-known as an important field in Computer Science due to the necessity of understanding the increasing complexity of Artificial Intelligence (AI) systems or algorithms. This is the reason why we can find a wide variety of explanation techniques (explainers) in the literature, on top of some XAI libraries. The challenge faced by XAI designers here is deciding what explainers are the most suitable for each scenario, taking into account the AI model, task to explain, user preferences, needs and knowledge, and overall, fitting into the explanation requirements. With the aim of addressing this problem, the iSee project was conceived to provide XAI design users with supporting tools to build their own explanation experiences. As a result, we have developed iSee, a Case-Based Reasoning-driven platform that allows users to create personalised explanation experiences. With the iSee platform, users add their explanation experience requirements, and get the most suitable XAI strategies to explain their own situation, taking advantage of XAI strategies previously used with success in similar context. The iSee platform is composed of different tools and modules: the ontology, the cockpit, the explainer library, the Explanation Experiences Editor (iSeeE3), the chatbot, and the analytics dashboard. This paper introduces these tools as a demo and tutorial for current and future users and for the XAI community. © 2024 Copyright for this paper by its authors.

关键词： Case based reasoning

来源：评论

学校读者我要写书评

暂无评论

Cost-Optimized Microservice Deployment for IoT Application in Cloud-Edge Collaborative Environment 26

Cost-Optimized Microservice Deployment for IoT Application i...

引用

26th International Conference on Computer Supported Cooperative Work in Design, CSCWD 2023

作者： Zhang, Xiaoyuan Tang, Bing Yang, Qing Xu, Wei Guo, Feiyan Hunan University of Science and Technology Hunan Key Laboratory for Service Computing and Novel Software Technology School of Computer Science and Engineering Xiangtan411201 China Guangzhou Maritime University School of Information and Communication Engineering Guangzhou510725 China

ISBN: (纸本)9798350331684

With the popularity of cloud native and DevOps, container technology is widely used and combined with microservices. The deployment of container-based microservices in distributed cloud-edge infrastructure requires suitable strategies to ensure the quality of service for users. However, the existing container orchestration tools cannot flexibly select the best deployment location according to the user's cost budget, and are insufficient in personalized deployment solutions. From the perspective of application providers, this paper considers the location distribution of users, application dependencies, and server price differences, and proposes a genetic algorithm-based Internet-of-Things (IoT) application deployment strategy for personalized cost budgets. The application deployment problem is defined as an optimization problem that minimizes user service latency under cost constraints. This problem is an NP-hard problem, and genetic algorithm is introduced to solve the optimization problem effectively and improve the deployment efficiency. The proposed algorithm is compared with four baseline algorithms, Time-Greedy, Cost-Greedy, Random and PSO, using real datasets and some synthetic datasets. The results show that the proposed algorithm outperforms other competing baseline algorithms. © 2023 IEEE.

关键词： Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

LOPE: A Low-Overhead Payment Verification Method for Blockchains

引用

Chinese Journal of Electronics 2021年第2期30卷 349-358页

作者： ZHANG Ying HE Zhengyuan JIA Linpeng DING Donghui ZHANG Hanwen SUN Yi School of Software Engineering Tongji University Blockchain Lab Institute of Computing Technology Chinese Academy of Sciences School of Computer Science and Technology University of Chinese Academy of Sciences

With its characteristics of decentralization,security, data traceability, and tamper-resistance, the blockchain has been widely used in various *** the difference in the performance of the devices, the light client is proposed so that devices without the ability to store a full blockchain copy can also participate in the blockchain transactions. However,the light client has to communicate with full nodes and verify the authenticity of a transaction which brings in some extent of communication, computation, and storage overheads to the light client. These overheads cannot be ignored for some low-performance devices, such as embedded devices or Io T chips, and therefore the current light client scheme does not work in this situation. We propose LOPE(a Low-overhead payment v Erification method) for poor-capacity nodes in the blockchain *** LOPE, a grouping protocol is designed to partition full nodes into groups to serve the verification requests of the light client. In addition, Practical byzantine fault tolerance(PBFT) is used to ensure the light client to get a credible result in spite of a few dishonest nodes existing in the group. We conduct LOPE and evaluate it in a testbed. The experiment results show that LOPE reduces more than half of the communication overhead, degrades the computation overhead of the light client to a large extent, and avoids the storage overhead of the hash roots of block headers in the light client. We also conduct theoretical analysis to show the performance improvement and security issues of LOPE.

关键词： current light client scheme storage overhead blockchains Low-overhead payment vErification method peer-to-peer computing software fault tolerance electronic money data integrity LOPE low-performance devices cryptographic protocols fault tolerance blockchain transactions

来源：评论

学校读者我要写书评

暂无评论

Analysis of Non-Fungible Token Pricing Factors with Machine Learning

Analysis of Non-Fungible Token Pricing Factors with Machine ...

引用

2022 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2022

作者： Ho, Kin-Hon Hou, Yun Chan, Tse-Tin Pan, Haoyuan The Hang Seng University of Hong Kong Department of Computing Hong Kong Shenzhen University College of Computer Science and Software Engineering China

ISBN: (数字)9781665452588

ISBN: (纸本)9781665452588

Rarity is known to be a factor in the price of non-fungible tokens (NFTs). Most investors make their purchasing decisions based on the rarity score or rarity rank of NFTs. However, not all rare NFTs are associated with a higher price, especially for play-to-earn gaming NFTs. In this paper, we studied the top-ranked play-to-earn gaming NFTs on Axie Infinity. We found that, in addition to rarity, utility is also a significant factor influencing the price. Furthermore, we use utility as a predictor to predict the price of Axies using the XGBoost regressor. Our results reveal that, compared to using rarity-based predictors only, leveraging utility-based predictors can improve the prediction accuracy, thus highlighting utility as a price determinant for play-to-earn gaming NFTs. © 2022 IEEE.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

The Robustness of Spiking Neural Networks in Communication and its Application towards Network Efficiency in Federated Learning

The Robustness of Spiking Neural Networks in Communication a...

引用

IEEE International Conference on Performance, computing and Communications (IPCCC)

作者： Manh V. Nguyen Liang Zhao Bobin Deng William Severa Honghui Xu Shaoen Wu College of Computing and Software Engineering Kennesaw State University Georgia USA Department of Cognitive & Emerging Computing Sandia National Laboratories New Mexico USA

ISBN: (数字)9798350367942

ISBN: (纸本)9798350367959

Spiking Neural Networks (SNNs) have recently gained significant interest in on-chip learning in embedded devices and emerged as an energy-efficient alternative to conventional Artificial Neural Networks (ANNs). However, to extend SNNs to a Federated Learning (FL) setting involving collaborative model training, the communication between the local devices and the remote server remains the bottleneck, which is often restricted and costly. In this paper, we first explore the inherent robustness of SNNs under noisy communication in FL. Building upon this foundation, we propose a novel Federated Learning with Top-κ Sparsification (FLTS) algorithm to reduce the bandwidth usage for FL training. We discover that the proposed scheme with SNNs allows more bandwidth savings compared to ANNs without impacting the model’s accuracy. Additionally, the number of parameters to be communicated can be reduced to as low as 6% of the size of the original model. We further improve the communication efficiency by enabling dynamic parameter compression during model training. Extensive experiment results demonstrate that our proposed algorithms significantly outperform the baselines in terms of communication cost and model accuracy and are promising for practical network-efficient FL with SNNs.

关键词： Training Accuracy Costs Federated learning Noise Spiking neural networks Bandwidth Robustness System-on-chip Noise measurement

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：