To overcome the limitation of the traditional clustering algorithms which fail to produce meaningful clusters in high-dimensional, sparseness and binary value data sets, a new method based on hypergraph model is propo...To overcome the limitation of the traditional clustering algorithms which fail to produce meaningful clusters in high-dimensional, sparseness and binary value data sets, a new method based on hypergraph model is proposed. The hypergraph model maps the relationship present in the original data in high dimensional space into a hypergraph. A hyperedge represents the similarity of attrlbute-value distribution between two points. A hypergraph partitioning algorithm is used to find a partitioning of the vertices such that the corresponding data items in each partition are highly related and the weight of the hyperedges cut by the partitioning is minimized. The quality of the clustering result can be evaluated by applying the intra-cluster singularity value. Analysis and experimental results have demonstrated that this approach is applicable and effective in wide ranging scheme.展开更多
This paper discusses the features and relevant theories of GIS spatial data model based on hypergraph,etc.The integrated concept model based on hypergraph and object_oriented model (HOOM) is proposed by the authors.Th...This paper discusses the features and relevant theories of GIS spatial data model based on hypergraph,etc.The integrated concept model based on hypergraph and object_oriented model (HOOM) is proposed by the authors.The principal contribution of this paper is that we study the K_section and other theories of hypergraph.An application example using HOOM is given at the end of the paper.展开更多
The rapid development of Internet technology makes it possible to integrate GIS with the Internet,forming Internet GIS.Internet GIS is based on a distributed client/server architecture and TCP/IP & IIOP.When const...The rapid development of Internet technology makes it possible to integrate GIS with the Internet,forming Internet GIS.Internet GIS is based on a distributed client/server architecture and TCP/IP & IIOP.When constructing and designing Internet GIS,we face the problem of how to express information units of Internet GIS.In order to solve this problem,this paper presents a distributed hypermap model for Internet GIS.This model provides a solution to organize and manage Internet GIS information units.It also illustrates relations between two information units and in an internal information unit both on clients and servers.On the basis of this model,the paper contributes to the expressions of hypermap relations and hypermap operations.The usage of this model is shown in the implementation of a prototype system.展开更多
Cloud storage has the characteristics of distributed and virtual, and it makes the ownership rights and management rights of users data separated. The master-slave architecture of cloud storage has a problem of single...Cloud storage has the characteristics of distributed and virtual, and it makes the ownership rights and management rights of users data separated. The master-slave architecture of cloud storage has a problem of single point failure. In this paper, we provide a cloud storage architecture model based on Semantic equivalence. According to semantic matching degree, this architecture divides the nodes into node cluster by creating semantic tree and maintains system routing through semantic hypergraph. Through simulation experiments show that dividing network into semantic can enhance scalability and flexibility of the system, and it can improve the efficiency of network organization and the security of cloud storage system, at the same time, it can also reduce the cloud data storage and the delay of reading time.展开更多
在许多KDD(knowledge discovery in databases)应用中,如电子商务中的欺诈行为监测,例外情况或离群点的发现比常规知识的发现更有意义.现有的离群点发现大多是针对数值属性的,而且这些方法只能发现离群点不能对其含义进行解释.提出了一...在许多KDD(knowledge discovery in databases)应用中,如电子商务中的欺诈行为监测,例外情况或离群点的发现比常规知识的发现更有意义.现有的离群点发现大多是针对数值属性的,而且这些方法只能发现离群点不能对其含义进行解释.提出了一种基于超图模型的离群点(outlier)定义,这一定义既体现了“局部”的概念能很好地解释离群点的含义.同时给出了HOT(hypergraph-based outlier test)算法,通过计算每个点的支持度、隶属度和规模偏差来检测离群点.该算法既能够处理数值属性,又能够处理类别属性.分析表明,该算法能有效地发现高维空间数据中的离群点.展开更多
文摘To overcome the limitation of the traditional clustering algorithms which fail to produce meaningful clusters in high-dimensional, sparseness and binary value data sets, a new method based on hypergraph model is proposed. The hypergraph model maps the relationship present in the original data in high dimensional space into a hypergraph. A hyperedge represents the similarity of attrlbute-value distribution between two points. A hypergraph partitioning algorithm is used to find a partitioning of the vertices such that the corresponding data items in each partition are highly related and the weight of the hyperedges cut by the partitioning is minimized. The quality of the clustering result can be evaluated by applying the intra-cluster singularity value. Analysis and experimental results have demonstrated that this approach is applicable and effective in wide ranging scheme.
文摘This paper discusses the features and relevant theories of GIS spatial data model based on hypergraph,etc.The integrated concept model based on hypergraph and object_oriented model (HOOM) is proposed by the authors.The principal contribution of this paper is that we study the K_section and other theories of hypergraph.An application example using HOOM is given at the end of the paper.
文摘The rapid development of Internet technology makes it possible to integrate GIS with the Internet,forming Internet GIS.Internet GIS is based on a distributed client/server architecture and TCP/IP & IIOP.When constructing and designing Internet GIS,we face the problem of how to express information units of Internet GIS.In order to solve this problem,this paper presents a distributed hypermap model for Internet GIS.This model provides a solution to organize and manage Internet GIS information units.It also illustrates relations between two information units and in an internal information unit both on clients and servers.On the basis of this model,the paper contributes to the expressions of hypermap relations and hypermap operations.The usage of this model is shown in the implementation of a prototype system.
基金supported in part by the National Science and technology support program of China No. 2014BAH29F05the National High-Tech R&D Program (863 Program) No. 2015AA01A705+3 种基金the National Natural Science Foundation of China under Grant No. 61572072the National Science and Technology Major Project No. 2015ZX03001041the Fundamental Research Funds for the Central Universities No. FRF-TP-14-046A2"Research on the System of Personalized Education using Big Data"
文摘Cloud storage has the characteristics of distributed and virtual, and it makes the ownership rights and management rights of users data separated. The master-slave architecture of cloud storage has a problem of single point failure. In this paper, we provide a cloud storage architecture model based on Semantic equivalence. According to semantic matching degree, this architecture divides the nodes into node cluster by creating semantic tree and maintains system routing through semantic hypergraph. Through simulation experiments show that dividing network into semantic can enhance scalability and flexibility of the system, and it can improve the efficiency of network organization and the security of cloud storage system, at the same time, it can also reduce the cloud data storage and the delay of reading time.
