Editors Recommend
15 January 2020, Volume 6 Issue 1
A high-dimensional numeric data collection algorithm for local difference privacy based on random projection
Huizhong SUN,Jianyu YANG,Xiang CHENG,Sen SU
2020, 6(1):  0.  doi:10.11959/j.issn.2096-0271.2020001
Asbtract ( 3 )   HTML ( 5)   PDF (1644KB) ( 4 )   Knowledge map   
Figures and Tables | References | Related Articles | Metrics

The problem of high-dimensional data collection satisfying local differential privacy was studied.A new locally differentially private algorithm called Multi-RPHM was proposed based on the random projection technology,which achieved the high utility of the collected high-dimensional numeric data while satisfying the local differential privacy.The algorithm was formally proved to meet ε-local differential privacy.The effectiveness of the algorithm was comfirmed through experiments on synthetic datasets.

Big data privacy protection based on secure compressive sensing
Ping WANG,Yushu ZHANG,Xing HE,Sheng ZHONG
2020, 6(1):  0.  doi:10.11959/j.issn.2096-0271.2020002
Asbtract ( 4 )   HTML ( 4)   PDF (1733KB) ( 6 )   Knowledge map   
Figures and Tables | References | Related Articles | Metrics

The current “big bang” of data is mainly driven by interconnection of all things.Various types of IoT sensing devices serving in daily life are constantly capturing personal privacy data.However,these privacy data have become the key targets of network attacks.Data security issues in the resource-constrained IoT applications were analyzed,a novel privacy protection technique based on compressive sensing theory was introduced,which is called secure compressed sensing,and a corresponding big data collection scheme was proposed.As demonstrated by the theoretical and experimental security analysis,there is a conclusive appeal for that secure compressive sensing can be used as a lightweight encryption mechanism which is built into the perception layer to provide first-level security protection for data at almost zero cost.

Same origin based fine-grained privacy protection for mobile applications
Wenxiong LU,Haoyu WANG
2020, 6(1):  0.  doi:10.11959/j.issn.2096-0271.2020003
Asbtract ( 1 )   HTML ( 0)   PDF (1533KB) ( 2 )   Knowledge map   
Figures and Tables | References | Related Articles | Metrics

Mobile systems,such as Android,use permission-based access control mechanism,which is at the granularity of each application.Apart from the code from developers themselves,applications also contain code from third-party libraries,which has led to serious overuse of application permissions.A novel origin-based (similar to browsers) and fine-grained control mechanism was introduced,which broke the boundary between applications in terms of access control and finegrained the granularity to the level of code source.The mechanism was implemented onto Android framework,and a set of tools to modify applications were also offered.Experiment results suggest that system can allow (or limit) certain developers to execute certain sensitive behaviors.

Data privacy,monopoly and fairness for AI
Xiaofeng MENG,Leixia WANG,Junxu LIU
2020, 6(1):  0.  doi:10.11959/j.issn.2096-0271.2020004
Asbtract ( 3 )   HTML ( 3)   PDF (1047KB) ( 6 )   Knowledge map   
Figures and Tables | References | Related Articles | Metrics

With the coming of the era of artificial intelligence,the value contained in big data has been deeply mined.But at the same time,the privacy and data monopoly issues of users’ sensitive data,and fairness in algorithmic decisions have become increasingly serious.In order to explore such problems,firstly,the development of data was researched,which reflects the unique producing environment of data ethics in the era of artificial intelligence,and the unique properties of these ethical issues were discussed.Then,the data monopoly,privacy disclosure and unfair decision-making were discussed one by one,whose development status and challenges were analyzed.It is concluded that the essence of current ethical issues is the non-transparency of data collection,data usage and algorithm decision,so that establishing the data transparency mechanism should be an important measure to solve these problems.

Research progress on risk analysis for artificial intelligence
Qun CHEN,Zhaoqiang CHEN,Boyi HOU,Lijuan WANG,Yuchen LUO,Zhanhuai LI
2020, 6(1):  0.  doi:10.11959/j.issn.2096-0271.2020005
Asbtract ( 6 )   HTML ( 0)   PDF (1453KB) ( 4 )   Knowledge map   
Figures and Tables | References | Related Articles | Metrics

The predictions of the deep learning models are still uncertain and uninterpretable.As a result,their deployments bring unavoidable risk to business decision making.Firstly,the study on risk analysis was motivated,and the three desirable properties of risk analysis techniques were described:quantifiability,interpretability and learnability.Then the existing work on risk analysis was reviewed,and the newly proposed framework to enable quantifiable,interpretable and learnable risk analysis was introduced.Finally,the existing and potential applications of risk analysis,and its future research direction were discussed.

Loss function and application research in supervised learning
Jianguo DENG,Sulan ZHANG,Jifu ZHANG,Yaling XUN,Aiqin LIU
2020, 6(1):  0.  doi:10.11959/j.issn.2096-0271.2020006
Asbtract ( 5 )   HTML ( 1)   PDF (1407KB) ( 17 )   Knowledge map   
Figures and Tables | References | Related Articles | Metrics

The loss function in supervised learning is often used to evaluate the degree of inconsistency between the real value of the sample and the predicted value of the model,and is generally used for parameter estimation of the model.Due to the constraints of application scenarios,data sets and problems to be solved,there are many kinds and quantities of loss functions used by existing supervised learning algorithms,and each loss function has its own characteristics.Therefore,it is very difficult to select a loss function suitable for solving the optimal model of the problem from many loss functions.The standard forms,basic ideas,advantages and disadvantages,main applications and corresponding evolution forms of commonly used loss functions in supervised learning algorithms were studied,and their more appropriate application scenarios and possible optimization strategies were summarized.This study not only helps to improve the accuracy of model prediction,it also provides a new idea for the application of new loss functions or to improve the application of existing loss functions.

A scheduler system for large-scale distributed data computing in cloud
Wanggen LIU,Huaicheng ZHENG,Guoping RONG
2020, 6(1):  0.  doi:10.11959/j.issn.2096-0271.2020007
Asbtract ( 7 )   HTML ( 3)   PDF (1543KB) ( 4 )   Knowledge map   
Figures and Tables | References | Related Articles | Metrics

A novel scheduler system including resource scheduling,application scheduling,configuration and label management center,cloud network and cloud storage services was introduced.The locality of computation and data was ensured by the ability of data topology awareness,and the I/O cost was saved.The impact of network storm was solved by optimizing the resource scheduling of point to point large data reading.The service level protocol was guaranteed by network and disk isolation technology and preemptive way.

Application research of big data technology in rural portrait
Wangyue LI,Jin LIU,Na CHEN
2020, 6(1):  0.  doi:10.11959/j.issn.2096-0271.2020008
Asbtract ( 2 )   HTML ( 1)   PDF (1543KB) ( 2 )   Knowledge map   
Figures and Tables | References | Related Articles | Metrics

Under the background of the national big data strategy and the rural revitalization strategy,big data technology was innovatively introduced into rural portrait.On the basis of systematically studying the research on rural big data sources,big data portrait technical level and the hotspots,key points and difficult problems of rural revitalization research,the conceptual model of the big data rural portraits were constructed from three aspects:rural development foundation,rural development status and rural development behavior.With the help of three types of labeling methods:original index,knowledge map and policy text,the conceptual model was concretized as three sets of label systems that can be used in practice,and on this basis,the big data calculation methods and visualization methods of the three sets of label systems were expounded,and taking examples of portraits.Finally,the future research direction of big data rural portrait was explored.

2019, Vol. 5 No.6 No.5 No.4 No.3 No.2 No.1
2018, Vol. 4 No.6 No.5 No.4 No.3 No.2 No.1
2017, Vol. 3 No.6 No.5 No.4 No.3 No.2 No.1
2016, Vol. 2 No.6 No.5 No.4 No.3 No.2 No.1
2015, Vol. 1 No.4 No.3 No.2 No.1
Most Download
Most Read
Most Cited