检索结果-南通市图书馆

crowdsourced federated learning architecture with personalized privacy preservation

Intelligent and Converged Networks 2024年第3期5卷 192-206页

作者： Yunfan Xu Xuesong Qiu Fan Zhang Jiakai Hao State Key Laboratory of Networking and Switching Technology Beijing University of Posts and TelecommunicationsBeijing 100876China State Grid Beijing Electric Power Company Beijing 100031China

In crowdsourced federated learning,differential privacy is commonly used to prevent the aggregation server from recovering training data from the models uploaded by clients to achieve privacy ***,improper privacy budget settings and perturbation methods will severely impact model *** order to achieve a harmonious equilibrium between privacy preservation and model performance,we propose a novel architecture for crowdsourced federated learning with personalized privacy *** our architecture,to avoid the issue of poor model performance due to excessive privacy preservation requirements,we establish a two-stage dynamic game between the task requestor and clients to formulate the optimal privacy preservation strategy,allowing each client to independently control privacy preservation ***,we design a differential privacy perturbation mechanism based on weight *** divides the weights based on their relevance with local data,applying different levels of perturbation to different types of ***,we conduct experiments on the proposed perturbation mechanism,and the experimental results indicate that our approach can achieve better global model performance with the same privacy budget.

关键词： crowdsourced federated learning differential privacy privacy preservation two-stage game

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Automatic test report augmentation to assist crowdsourced testing

引用

Frontiers of Computer Science 2019年第5期13卷 943-959页

作者： Xin CHEN He JIANG Zhenyu CHEN Tieke HE Liming NIE School of Computer Science and Technology Hangzhou Dianzi University Hangzhou 310018 China School of Software Dalian University of Technology Dalian 116621 China Key Laboratory for Ubiquitous Network and Service Software of Liaoning Province Dalian 116621 China School of Software Nanjing University Nanjing 210093 China School of Information Science and Technology Zhejiang Sci-Tech University Hangzhou 310018 China

In crowdsourced mobile application testing, workers are often inexperienced in and unfamiliar with software testing. Meanwhile, workers edit test reports in descriptive natural language on mobile devices. Thus, these test reports generally lack important details and challenge developers in understanding the bugs. To improve the quality of inspected test reports, we issue a new problem of test report augmentation by leveraging the additional useful information contained in duplicate test reports. In this paper, we propose a new framework named test report augmentation framework (TRAF) towards resolving the problem. First, natural language processing (NLP) techniques are adopted to preprocess the crowdsourced test reports. Then, three strategies are proposed to augment the environments, inputs, and descriptions of the inspected test reports, respectively. Finally, we visualize the augmented test reports to help developers distinguish the added information. To evaluate TRAF, we conduct experiments over five industrial datasets with 757 crowdsourced test reports. Experimental results show that TRAF can recommend relevant inputs to augment the inspected test reports with 98.49% in terms of NDCG and 88.65% in terms of precision on average, and identify valuable sentences from the descriptions of duplicates to augment the inspected test reports with 83.58% in terms of precision, 77.76% in terms of recall, and 78.72% in terms of F-measure on average. Meanwhile, empirical evaluation also demonstrates that augmented test reports can help developers understand and fix bugs better.

关键词： crowdsourced testing test report TF-IDF natural language processing test report augmentation

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

An accurate indoor map matching algorithm based on activity detection and crowdsourced Wi-Fi

引用

Science China(Technological Sciences) 2019年第9期62卷 1492-1501页

作者： YU WenPing ZHANG JianZhong XU JingDong XU YuWei College of Computer Science Nankai University

Map matching has been widely investigated in indoor pedestrian navigation to improve positioning accuracy and *** paper proposes an accurate map matching algorithm based on activity detection and crowdsourced Wi-Fi(AiFiMatch).Firstly, by taking indoor road segments between activity-related locations as nodes, and the activity type from one road segment to another as directed edge, the indoor floor plan is abstracted as a directed graph. Secondly, the smartphone’s motion sensors are utilized to detect different activities based on a decision tree and then the pedestrian’s walking trajectory is divided into subtrajectory sequence according to location-related activities. Finally, the sub-trajectory sequence is matched to the directed graph of indoor floor plan to position the pedestrian by using a Hidden Markov Model(HMM). Simultaneously, Wi-Fi fingerprints are bound to road segments based on timestamp. Through crowdsourcing, a radio map of indoor road segments is constructed. The radio map in turn inversely promotes the HMM based map matching algorithm. AiFiMatch is evaluated by the experiments using smartphones in a teaching building. Experimental results show that the pedestrian can be accurately tracked even without knowing the starting position and AiFiMatch is robust to a certain degree of step length and heading direction errors.

关键词： map matching hidden markov model activity detection crowdsourced Wi-Fi

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Harnessing crowdsourced Data and Prevalent Technologies for Atmospheric Research

引用

Advances in Atmospheric Sciences 2019年第7期36卷 766-769页

作者： Noam DAVID Institute of Industrial Science The University of Tokyo

The knowledge garnered in environmental science takes a crucial part in informing decision-making in various fields,including agriculture, transportation, energy, public health and safety, and more. Understanding the basic processes in each of these fields relies greatly on progress being made in conceptual, observational and technological approaches. However,existing instruments for environmental observations are often limited as a result of technical and practical constraints. Current technologies, including remote sensing systems and ground-level measuring means, may suffer from obstacles such as low spatial representativity or a lack of precision when measuring near ground-level. These constraints often limit the ability to carry out extensive meteorological observations and, as a result, the capacity to deepen the existing understanding of atmospheric phenomena and processes. Multi-system informatics and sensing technology have become increasingly distributed as they are embedded into our environment. As they become more widely deployed, these technologies create unprecedented data streams with extraordinary levels of coverage and immediacy, providing a growing opportunity to complement traditional observation techniques using the large volumes of data created. Commercial microwave links that comprise the data transfer infrastructure of cellular communication networks are an example of these types of systems. This viewpoint letter briefly reviews various works on the subject and presents aspects concerning the added value that may be obtained as a result of the integration of these new means, which are becoming available for the first time in this era, for studying and monitoring atmospheric phenomena.

关键词： atmospheric science IoT(Internet of Things) crowdsourced data commercial microwave links

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案：

请选择收藏分类：

通借通还

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

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：