检索结果-内蒙古大学图书馆

International Journal of Digital Content technology and its Applications 2012年第22期6卷 580-589页

作者： Ren, Jiadong Liu, Zhigang Dong, Jun College of Information Science and Engineering Yanshan University Qinhuangdao Hebei 066004 China Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province Qinhuangdao City 066004 China

In high-dimensional data space, because the data is sparse inherently, clusters tend to exist in different subspaces, which makes the traditional methods no longer suitable for use. In this paper, we present SCFES, a subspace clustering algorithm based on finding effective spaces. First, we define the effective dimension. By calculating relative entropy we remove redundancy dimensions which affect clustering accuracy. Second, according to the data distribution in the effective dimensions, we get the effective intervals through merging adjacent intervals. The effective space is composed of effective intervals. Third, we extend the density estimator based on undirected acyclic connected graph by using weight so as to estimate the expectation of existing clusters in the space, at the same time combine it with the monotonicity of the clustering criterion mentioned in the CLIQUE algorithm to prune candidates. Consequently we get the effective spaces. Finally, we adopt the structure of sibling tree to store all the effective spaces and use DBSCAN algorithm based on density to generate maximal subspace clusters in some effective spaces. Experimental results show that SCFES effectively finds arbitrarily shaped and positioned clusters in different subspaces. Meanwhile SCFES has better clustering quality and scalability.

关键词： Clustering algorithms

来源：评论

学校读者我要写书评

暂无评论

IMSPMIS-Stream: incremental mining of top-k short sequential pattern over multiple item set streams

引用

Journal of Computational Information systems 2014年第6期10卷 2305-2312页

作者： Hao, Xiaobing Han, Gaowei Chen, Yuping Wang, Peilong Ren, Jiadong College of Information Science and Engineering Yanshan University Qinhuangdao 066004 China The Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province Qinhuangdao 066004 China

Many previous algorithms in data streams are about single stream, which can only process single items. The algorithms about data streams are always extended by sequential pattern algorithms about static database, they can't satisfy the requirement of scanning streams only once and online mining. Moreover, top-k short sequential pattern hasn't been studied yet. Therefore, IMSPMIS-Stream algorithm is proposed to mine online multiple streams with the single scan by the incremental method in this paper. The streams are continuous transactions, which is also called item set streams. Several streams are mined interactively based on sliding windows at the same time. An approach to count the support of every sub item set is designed to get the support during one scan. The short sequential pattern is deffned, whose item sets are shorter than all the item sets with the same occurring time. When the arrival transactions reach to the capacity of the window, sequential patterns will be output by combining top-k short item sets. The experiments show that the IMSPMIS-Stream algorithm overcomes the memory limitation and its execution time is reduced compared with some previous algorithms under the same conditions. Copyright © 2014 Binary Information Press.

关键词： Information systems

来源：评论

学校读者我要写书评

暂无评论

A Robust Collaborative Recommendation Algorithm Based on k-distance and Tukey M-estimator

引用

China Communications 2014年第9期11卷 112-123页

作者： YI Huawei ZHANG Fuzhi LAN Jie School of Information Science and Engineering Yanshan University Qinhuangdao 066004 Hebei Province E R.China The Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province Qinhuangdao 066004 P. R. China Liaoning University of Technology Jinzhou 121001 Liaoning Province P. R. China

The existing collaborative recommendation algorithms have lower robustness against shilling *** this problem in mind,in this paper we propose a robust collaborative recommendation algorithm based on k-distance and Tukey ***,we propose a k-distancebased method to compute user suspicion degree(USD).The reliable neighbor model can be constructed through incorporating the user suspicion degree into user neighbor *** influence of attack profiles on the recommendation results is reduced through adjusting similarities among ***,Tukey M-estimator is introduced to construct robust matrix factorization model,which can realize the robust estimation of user feature matrix and item feature matrix and reduce the influence of attack profiles on item feature ***,a robust collaborative recommendation algorithm is devised by combining the reliable neighbor model and robust matrix factorization *** results show that the proposed algorithm outperforms the existing methods in terms of both recommendation accuracy and robustness.

关键词： shilling attacks robust collaborative recommendation matrix factori-zation k-distance Tukey M-estimator

来源：评论

学校读者我要写书评

暂无评论

Mining time-interval weighted closed sequential patterns based on memory indexing

引用

Journal of Computational Information systems 2014年第1期10卷 293-300页

作者： Zeng, Qiang Chen, Dengxi Ren, Jiadong Han, Gaowei College of Information Science and Engineering Yanshan University Qinhuangdao 066004 China The Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province Qinhuangdao 066004 China

General weighted sequential pattern mining algorithms ignore or do not make good use of the time and time-interval information of data elements. Besides some algorithms require to scan the database many times or build temporary databases. To solve these problems, we propose a memory-based algorithm MITWCSpan (Memory Indexing for time-interval Weighted Closed Sequential pattern mining) for timeinterval weighted closed sequential pattern mining. The algorithm takes full account of the importance of the time-interval of data elements. Moreover, an improved index set based on time-interval, p-tidx, is defined. During mining process, the algorithm adopts the find-then-index technique recursively to find the items which can constitute a time-interval weighted sequential pattern and construct p-tidx for the possible sequential pattern. Finally the algorithm uses closing detection to get the whole time-interva l weighted closed sequential patterns. The experimental results show that the algorithm is more efficient in finding more important sequential patterns. Copyright © 2014 Binary Information Press.

关键词： Indexing (of information)

来源：评论

学校读者我要写书评

暂无评论

A density grid-based uncertain data stream clustering algorithm

引用

Journal of Computational Information systems 2014年第9期10卷 3619-3626页

作者： He, Haitao Zhao, Jintian The Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province Qinhuangdao 066004 China College of Information Science and Engineering Yanshan University Qinhuangdao 066004 China

The existing grid-based uncertain data stream clustering algorithms are fast but low-accuracy, and sensitive to user-specified threshold. In order to solve the above problems, a density grid-based uncertain data stream clustering algorithm UG-Stream is proposed in this paper. In UG-Stream algorithm, a dynamic threshold is defined by taking uncertainty and grid feature into account, dense grid can be distinguished by the threshold. The probability variance is defined to describe the distribution of internal data points in grid. If grid distribution can be taken as uniform, dense grid can be classified as core dense grid by probability variance. Core dense grid can be clustered directly, the rest of dense grids will be merged into current existing clusters by probability center distance. Contrast experiments show that UG-Stream algorithm is superior to UMicro algorithm in both clustering accuracy and clustering rate. © 2014 Binary Information Press.

关键词： Clustering algorithms

来源：评论

学校读者我要写书评

暂无评论

Density-based clustering for evolving uncertain data stream

引用

Journal of Computational Information systems 2014年第1期10卷 419-426页

The current clustering algorithms for evolving uncertain data stream are sensitive to user specified threshold, and unstable in noise processing. In this paper, DUStream is presented, a density-based algorithm for discovering clusters in evolving uncertain data stream. Probability distance is introduced as a new similarity measure, giving consideration to probability attribute and distance attribute. Probability Radius is used as a self-adaption dynamic threshold to reduce the effect of user specified input. The experimental results demonstrate the effectiveness and Effciency of the algorithm on Artificial and real data sets. Copyright © 2014 Binary Information Press.

关键词： Probability

来源：评论

学校读者我要写书评

暂无评论

An energy-efficient routing algorithm of zigbee network

引用

Journal of Computational Information systems 2012年第23期8卷 9873-9879页

作者： Liu, Yongshan Chen, Wei Shang, Xuehui Lv, Yongteng Han, Yuanyuan Liu, Chang College of Information Science and Engineering Yanshan University Qinhuangdao 066004 China The Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province Qinhuangdao 066004 China

As regard to the case of extending the lifetime of zigbee network, the defination of node's boundary is proposed. First, all the information for node's boundary is stored when zigbee network is built. Then, the packet between the nodes which are in the node's boundary is transferred directly. The packet between the nodes whose boundaries are adjacent is transferred through the node with maximum energy. The exprimental results show that the new algorithm can efficiently extend the lifetime of zigbee network. ©2012 by Binary Information Press.

关键词： Zigbee

来源：评论

学校读者我要写书评

暂无评论

virtual grid-based clustering of uncertain data on vulnerability database

引用

Journal of Convergence Information technology 2012年第20期7卷 429-438页

作者： Dong, Jun Cao, Mengmeng Huang, Guoyan Ren, Jiadong College of Information Science and Engineering Yanshan University Qinhuangdao 066004 China The Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province Qinhuangdao 066004 China

Most existing vulnerability taxonomy classifies vulnerabilities by their idiosyncrasies, weaknesses, flaws and faults et al. The disadvantage of the taxonomy is that the classification standard is not unified and there is overlap classification phenomenon in vulnerability taxonomy. In order to solve the problem, we will propose an algorithm VUNClique, virtual Grid-based Clustering of Uncertain Data on vulnerability database. Firstly, this paper transforms the vulnerability database into uncertain dataset using the existing vulnerability database pretreatment model. Secondly, we define a virtual grid structure, the cells are divided into real cells and virtual cells, but only the real cells which contain data objects stored in memory. The probability attribute value similarity is defined to deal with the similarity of non-numeric attributes, which compares the number of non-numeric attributes with the same value between tuples to measure the similarity. We provide a secondary partition algorithm to improve the similarity between the tuples in the same cell, the algorithm merges a tuple into it's high-density neighbor cell which has the maximum value of probability attribute value similarity with it. Then, a novel identify cluster algorithm is provided to cluster the high-density real cells. It can identify clusters of arbitrary shapes by traversing real cells twice. Finally, performance experiments over the uncertain dataset transformed by NVD vulnerability database. The experiments results show that VUNClique can find clusters of arbitrary shapes, and greatly improve the efficiency of clustering.

关键词： Cells

来源：评论

学校读者我要写书评

暂无评论

Robust recommendation algorithm based on user rating matrix block and modified LTS-estimator

引用

Journal of Information and Computational Science 2014年第6期11卷 1889-1898页

作者： Xu, Yuchen Liu, Zhen Zhang, Fuzhi School of Information Science and Engineering Yanshan University Qinhuangdao 066004 China The Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province Qinhuangdao 066004 China

The most widely-used collaborative recommendation algorithms are vulnerable to shilling attacks. To this end, in this paper we propose a robust recommendation algorithm based on user rating matrix block and modified LTS-estimator. Firstly, we construct user rating matrix blocks using user rating matrix block algorithm based on k-median clustering. Secondly, we apply the modified LTS-estimator to matrix factorization model in order to produce user feature matrix and item feature matrix. Finally, we devise a robust recommendation algorithm to generate recommendations for the target users. Experimental results on the MovieLens dataset show that the proposed algorithm outperforms the existing methods in terms of both the prediction accuracy and robustness. © 2014 Binary Information Press.

关键词： Matrix factorization

来源：评论

学校读者我要写书评

暂无评论

WCSPMPD-stream: Mining weighted closed sequential patterns with pattern decay over data streams

引用

Journal of Computational Information systems 2014年第1期10卷 435-442页

作者： Zeng, Qiang Han, Gaowei Li, Weina Ren, Jiadong College of Information Science and Engineering Yanshan University Qinhuangdao 066004 China The Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province Qinhuangdao 066004 China

Many of the previous incremental methods in data streams are deleting the old patterns and adding to the new patterns directly, which may delete useful patterns too early. Both different real data and the data occurring time can lead to the diverse importance, which are not considered at the same time in previous studies. Therefore, WCSPMPD-Stream, a weighted closed sequential pattern algorithm with pattern decay based on sliding windows over data streams is proposed in this paper. Firstly, the sliding windows are separated into several fragments to be mined respectively. Secondly, Time Weight Decay and No Updating Decay methods are designed to control the decay rate. Thirdly, a method of calculating the time weight is defined which varies linearly with the length of time interval between each stream fragment and the current stream fragment. Fourthly, an approach of getting the item weight is defined to calculate any sequence weight by the items in it. Fifthly, a new data structure WFPS-Tree is advanced to store and update the mining results. The experiments show that WCSPMPD-Stream algorithm overcomes the memory limitation and mines patterns with higher interest. And its execution time is reduced compared with the previous algorithms under the same conditions. Copyright © 2014 Binary Information Press.

关键词： Trees (mathematics)

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：