检索结果-内蒙古大学图书馆

Lecture Notes in Computer Science

作者： Liao, Guoqiong Wu, Linqing Wan, Changxuan Xiong, Naixue School of Information Technology Jiangxi University of Finance and Economics Nanchang 330013 China Jiangxi Key Laboratory of Data and Knowledge Engineering Nanchang 330013 China Department of Computer Science Georgia State University United States

ISBN: (纸本)9783642236402

In recent years, large amounts of uncertain data are emerged with the widespread employment of the new technologies, such as wireless sensor networks, RFID and privacy protection. According to the features of the uncertain data streams such as incomplete, full of noisy, non-uniform and mutable, this paper presents a probability frequent pattern tree called PFP-tree and a method called PFP-growth, to mine probability frequent patterns based on probability damped windows. The main characteristics of the suggested method include: (1) adopting time-based probability damped window model to enhance the accuracy of mined frequent patterns;(2) setting an item index table and a transaction index table to speed up retrieval on the PFP-tree;and (3) pruning the tree to remove the items that cannot become frequent patterns;. The experimental results demonstrate that PFP-growth method has better performance than the main existing schemes in terms of accuracy, processing time and storage space. © 2011 Springer-Verlag.

关键词： Probability

来源：评论

学校读者我要写书评

暂无评论

Selection of multimedia data source based on user feedback

Selection of multimedia data source based on user feedback

引用

2011 International Conference on Management of e-Commerce and e-Government, ICMeCG 2011

作者： Deng, Song Wan, Changxuan Liu, Xiping School of Information Technology Jiangxi University of Finance and Ecomomics Nanchang China School of Software Jiangxi Agricultural University Nanchang China Jiangxi Key Laboratory of Data and Knowledge Engineering Jiangxi University of Finance and Economics Nanchang China

ISBN: (纸本)9780769545448

There are hundreds or thousands of web data sources providing data of relevance to a particular domain on the Web, so how to find a suitable set of sources quickly to integrate from a number of sources is becoming more and more important. Traditional data sources selection methods that are based on data quality do not consider quality characteristics of data sources under the different domains, but select the same quality dimensions for quality evaluation models. As a result, accuracies of a traditional method under different domains are quite different. In light of this, we propose a Web multimedia source selection method based on user feedback. We gain accurate core quality dimensions of data sources under the multimedia domain based on user feedback, and then achieve a quality estimation model of the multimedia data sources. The experiments on hundreds of real-world multimedia data sources have proved that our approach is efficient and accurate. © 2011 IEEE.

关键词： Accuracy Computers data Source data models databases Equations Mathematical model Multimedia Multimedia communication Quality Dimensions User Feedback Web data source data quality estimation model quality evaluation model quality multimedia computing mult

来源：评论

学校读者我要写书评

暂无评论

K-nearest neighbors in uncertain graph

引用

Jisuanji Yanjiu yu Fazhan/Computer Research and Development 2011年第10期48卷 1850-1858页

作者： Zhang, Yinglong Li, Cuiping Chen, Hong Du, Lingxia Key Laboratory of Data Engineering and Knowledge Engineering Renmin University of China Ministry of Education Beijing 100872 China School of Computer and Information Engineer Jiangxi Agriculture University Nanchang 330045 China

In many areas, a lot of data have been modeled by graphs which are subject to uncertainties, such as molecular compounds and protein interaction networks. While many real applications, for example, collaborative filtering, fraud detection, and link prediction in social networks etc, rely on efficiently answering k-nearest neighbor queries (kNN), which is the problem of computing the most "similar" k nodes to a given query node. To solve the problem, in this paper a novel method based on measurement of SimRank is proposed. However, because graphs evolve over time and are uncertainly, the computing cost can be very high in practice to solve the problem using the existing algorithms of SimRank. So the paper presents an optimization algorithm. Introducing path threshold, which is suitable in both determined graph and uncertain graph, the algorithm merely considers the local neighborhood of a given query node instead of whole graph to prune the search space. To further improving efficiency, the algorithm adopts sample technology in uncertain graph. At the same time, theory and experiments interpret and verify that the optimization algorithm is efficient and effective.

关键词： Collaborative filtering

来源：评论

学校读者我要写书评

暂无评论

On Mutual Support of Modern and Traditional Access Control Models with UCON and BLP as Case Study

On Mutual Support of Modern and Traditional Access Control M...

引用

WRI International Conference on Communications and Mobile Computing, CMC

作者： Hui Feng Wenchang Shi Zhaohui Liang Bin Liang School of Information Key Laboratory of Data Engineering and Knowledge Engineering Ministry of Education Renmin University of China Beijing China

Access control is essential to computer security, especially in an open, distributed, networked communication environment. Modern access control model such as UCON aims at accommodating general requirements. Traditional one such as BLP focuses on specific properties, e.g. confidentiality. Both of these two realms have their limitations. Taking UCON and BLP as case study, this paper explores mutual support of modern and traditional access control models. It investigates BLP's adaptable characteristic in the UCON perspective. First, it constructs properties in the UCON language to manifest the BLP adaptability, which shows that the BLP adaptability can be ensured to function correctly by the UCON framework. Further, it proposes a formal specification for the BLP adaptability under the UCON framework with the Temporal Logic of Actions, which demonstrates that the BLP adaptability is in good consistency with the UCON model. The significance of the paper is twofold. On the one hand, it exhibits that adaptable quality of the traditional BLP model may be ensured theoretically by the philosophy of modern access control. On the other hand, it enriches the real sense of modern access control models by strengthening the power of traditional access control models.

关键词： Adaptation model Sensitivity Authorization Permission Computational modeling

来源：评论

学校读者我要写书评

暂无评论

图形处理器加速的联机分析处理系统

图形处理器加速的联机分析处理系统

引用

第29届中国数据库学术会议

作者： Fang Yixuan 方艺璇刘虹 Liu Hong Chen Hong 陈红 Li Cuiping 李翠平 Zhang Yansong 张延松 Zhao Suyun 赵素云 Chen Jie 陈杰辛鑫 Xin Xin Zhang Ji 张吉 Key Laboratory of Data Engineering and Knowledge Engineering (Renmin University of China) Ministry 数据工程与知识工程教育部重点实验室(中国人民大学) 北京100872 中国人民大学信息学院北京100872 Key Laboratory of Data Engineering and Knowledge Engineering (Renmin University of China) Ministry 数据工程与知识工程教育部重点实验室(中国人民大学) 北京100872 中国人民大学中国调查与数据中心北京100872

基于现有联机分析处理系统(online analytical processing,OLAP)的不足和图形处理器(graphics processing unit,GPU)的发展,研制了GOOLAP(GPU oriented OLAP)系统.GOOLAP系统利用GPU的高并行性和高存储带宽,把计算密集型运算转移到GPU端执行,加运OLAP处理性能.GOOLAP系统主要由3部分组成：1)表示层,以Excel作为用户输入前端并展示结果,加强系统的易用性;2)OLAP服务器层,负责把多维表达式(multi-dimensional expression,MDX)语句解析成SQL语句并根据自定义启发式规则向GPU传递适合计算的SQL语句,最后再将GPU返回结果与CPU原始结果合并;3)存储层,提供GPU显存和关系数据库两种存储结构.

关键词：图形处理器联机分析处理系统优化设计功能模块

来源：评论

学校读者我要写书评

暂无评论

MyBUD自适应分布式存储管理的设计与实现

MyBUD自适应分布式存储管理的设计与实现

引用

第29届中国数据库学术会议

作者： ZHOU Ningnan 周宁南 ZHANG Xiao 张孝 SUN Xinyun 孙新云 JU Xingxing 琚星星 LIU Kuicheng 刘奎呈 DU Xiaoyong 杜小勇 WANG Shan 王珊 Key Laboratory of Data Engineering and Knowledge Engineering Ministry of Education Renmin Universi 中国人民大学数据工程与知识工程教育部重点实验室北京100872 中国人民大学信息学院北京100872

面对日益增长的非结构化数据管理需求，实现了基于“自由表”数据模型和BUD参考体系模型的非结构化数据管理平台MyBUD系统。提出了一种能够根据非结构化数据的类型和访问特点自适应地选择分布式存储子系统的方法，同时也对MyBUD进行了T... 详细信息

面对日益增长的非结构化数据管理需求，实现了基于“自由表”数据模型和BUD参考体系模型的非结构化数据管理平台MyBUD系统。提出了一种能够根据非结构化数据的类型和访问特点自适应地选择分布式存储子系统的方法，同时也对MyBUD进行了TPCC测试和非结构化数据存取实验。结果表明，这种自适应的数据存储方法为MyBUD系统提供了高效的可扩展存储层，为采用数据库方法实现对结构化和非结构化数据统一管理的进一步研究工作奠定了基础。

关键词：数据库系统存储管理自适应算法体系架构服务质量

来源：评论

学校读者我要写书评

暂无评论

Analysis of bandwagon and average hybrid attack model against trust-based recommender systems

Analysis of bandwagon and average hybrid attack model agains...

引用

2011 International Conference on Management of e-Commerce and e-Government, ICMeCG 2011

作者： Zhang, Fuguo School of Information and Technology Jiangxi University of Finance and Economics Nanchang China Institute of Information Resource Management Jiangxi University of Finance and Economics Nanchang China Jiangxi Key Laboratory of Data and Knowledge Engineering Jiangxi University of Finance and Economics Nanchang China

ISBN: (纸本)9780769545448

Recommender systems have been accepted as a vital application on the web by offering product advice or information that users might be interested in. Despite its success, similarity-based collaborative filtering suffers from some significant limitations, such as scalability, sparsity and recommendation attack. Prior work has shown incorporating trust mechanism into traditional collaborative filtering recommender systems can improve these limitations. However, trust-based recommender systems are also known to be vulnerable to profile infection attacks. Malicious users can inject a large number of biased profiles into such a system in order to make recommendations that favor or disfavor given items. In this paper, we propose a bandwagon and average hybrid attack model and analysis the effectiveness of the attack model against topic-level trust-based recommender algorithm. The results of our experiments conducted on well-known dataset show that the hybrid attack model is more effective than other attack models. © 2011 IEEE.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

Sentiment classification via L2-norm deep belief network 11

Sentiment classification via L2-norm deep belief network

引用

20th ACM Conference on Information and knowledge Management, CIKM'11

作者： Liu, Tao Li, Minghui Zhou, Shusen Du, Xiaoyong Key Laboratory of Data Engineering and Knowledge Engineering MOE 100872 Beijing China School of Information Renmin University of China 100872 Beijing China Microsoft Asian Research and Development Group 100080 Beijing China Shenzhen Graduate School Harbin Institute of Technology 518055 Shenzhen China

ISBN: (纸本)9781450307178

Automatic analysis of sentiments expressed in large scale online reviews is very important for intelligent business applications. Sentiment classification is the most popular task of sentiment analysis, which is more challenging than traditional topic-based text classification. Basic features, such as vocabulary words, are not enough to classify sentiments well. Deep Belief Network (DBN) is introduced to discover more abstract features of sentiments. To capture full information of the features, large-size network can be constructed, but at the same time, large-size network tends to over fit the training data and even noise, which will reduce the generalization ability of the network. In this paper, L2-norm Deep Belief Network (L2DBN) is proposed, which uses L2-norm regularization to optimize the network parameters of DBN. L2DBN is first initialized by an unsupervised layer-wise training algorithm, and then fine-tuned by a supervised procedure. Network parameters are optimized using both classification loss and network complexity. Experimental results show that the proposed L2DBN outperforms the state-of-the-art method and the basic DBN on golden, noisy and heterogeneous datasets. © 2011 ACM.

关键词： Sentiment analysis

来源：评论

学校读者我要写书评

暂无评论

Lawyer information integration and recommendation by multi-source information validation

Lawyer information integration and recommendation by multi-s...

引用

International Conference on Machine Learning and Cybernetics (ICMLC)

作者： Tao Liu Biao Fan He Hu Xiao-Yong Du Key Laboratory of Data Engineering and Knowledge Engineering MOE Renmin University of China China School of Information Renmin University of China Beijing China

On the internet, all-round lawyer information is located at separated information sources, which prevent web users from effective information acquisition. In order to build a unified view of separated, heterogeneous, and often redundant lawyer information, we propose a new information integration method using multi-source information cross-validation. Based on the unified integrated data, a lawyer recommendation system is built. Several key technologies are presented and evaluated, including the multi-source information acquisition and validation. Experimental results indicate the key techniques used in the system are effective for lawyer information integration and recommendation.

关键词： Organizations Crawlers Search engines data mining Web pages databases Reliability

来源：评论

学校读者我要写书评

暂无评论

Parallel Aggregation Queries over Star Schema: A Hierarchical Encoding Scheme and Efficient Percentile Computing as a Case

Parallel Aggregation Queries over Star Schema: A Hierarchica...

引用

International Symposium on Parallel and Distributed Processing with Applications, ISPA

作者： Xiongpai Qin Huiju Wang Xiaoyong Du Shan Wang Ministry of Education Key Laboratory of Data Engineering and Knowledge Engineering RUC Beijing China School of Information Renmin University of China Beijing China

Big data analysis is a main challenge we meet recently. Cloud computing is attracting more and more big data analysis applications, due to its well scalability and fault-tolerance. Some aggregation functions, like SUM, can be computed in parallel, because they satisfy distributive law of addition. Unfortunately, some of statistical functions are not naturally parallelizable. That means they do not satisfy distributive law of addition. In this paper, we focus on percentile computing problem. We proposed an iterative-style prediction-based parallel algorithm in a distributed system. Prediction is done through a sampling technique. Experiment results verify the efficiency of our algorithm.

关键词： Prediction algorithms Convergence Algorithm design and analysis Histograms Encoding Query processing Indexes

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：