Results 1  10
of
38
Survey of clustering algorithms
 IEEE TRANSACTIONS ON NEURAL NETWORKS
, 2005
"... Data analysis plays an indispensable role for understanding various phenomena. Cluster analysis, primitive exploration with little or no prior knowledge, consists of research developed across a wide variety of communities. The diversity, on one hand, equips us with many tools. On the other hand, the ..."
Abstract

Cited by 429 (4 self)
 Add to MetaCart
(Show Context)
Data analysis plays an indispensable role for understanding various phenomena. Cluster analysis, primitive exploration with little or no prior knowledge, consists of research developed across a wide variety of communities. The diversity, on one hand, equips us with many tools. On the other hand, the profusion of options causes confusion. We survey clustering algorithms for data sets appearing in statistics, computer science, and machine learning, and illustrate their applications in some benchmark data sets, the traveling salesman problem, and bioinformatics, a new field attracting intensive efforts. Several tightly related topics, proximity measure, and cluster validation, are also discussed.
Revisiting T. Uno and M. Yagiura’s Algorithm
, 2005
"... ... of two given permutations of length n in O(n + K) time. Our paper first presents a decomposition approach to obtain a compact encoding for common intervals of d permutations. Then, we revisit T. Uno and M. Yagiura’s algorithm to yield a linear time algorithm for finding this encoding. Besides, w ..."
Abstract

Cited by 26 (8 self)
 Add to MetaCart
(Show Context)
... of two given permutations of length n in O(n + K) time. Our paper first presents a decomposition approach to obtain a compact encoding for common intervals of d permutations. Then, we revisit T. Uno and M. Yagiura’s algorithm to yield a linear time algorithm for finding this encoding. Besides, we adapt the algorithm to obtain a linear time modular decomposition of an undirected graph, and thereby propose a formal invariantbased proof for all these algorithms.
Drawing graphs using modular decomposition
 IN GRAPH DRAWING. VOLUME LNCS 3843
, 2005
"... In this paper we present an algorithm for drawing an undirected graph G which takes advantage of the structure of the modular decomposition tree of G. Specifically, our algorithm works by traversing the modular decomposition tree of the input graph G on n vertices and m edges, in a bottomup fashi ..."
Abstract

Cited by 12 (2 self)
 Add to MetaCart
(Show Context)
In this paper we present an algorithm for drawing an undirected graph G which takes advantage of the structure of the modular decomposition tree of G. Specifically, our algorithm works by traversing the modular decomposition tree of the input graph G on n vertices and m edges, in a bottomup fashion until it reaches the root of the tree, while at the same time intermediate drawings are computed. In order to achieve aesthetically pleasing results, we use grid and circular placement techniques, and utilize an appropriate modification of a wellknown spring embedder algorithm. It turns out, that for some classes of graphs, our algorithm runs in O(n+m) time, while in general, the running time is bounded in terms of the processing time of the spring embedder algorithm. The result is a drawing that reveals the structure of the graph G and preserves certain aesthetic criteria.
A Survey On: Content Based Image Retrieval Systems Using Clustering Techniques For Large Data sets
"... Contentbased image retrieval (CBIR) is a new but widely adopted method for finding images from vast and unannotated image databases. As the network and development of multimedia technologies are becoming more popular, users are not satisfied with the traditional information retrieval techniques. So ..."
Abstract

Cited by 11 (0 self)
 Add to MetaCart
(Show Context)
Contentbased image retrieval (CBIR) is a new but widely adopted method for finding images from vast and unannotated image databases. As the network and development of multimedia technologies are becoming more popular, users are not satisfied with the traditional information retrieval techniques. So nowadays the content based image retrieval (CBIR) are becoming a source of exact and fast retrieval. In recent years, a variety of techniques have been developed to improve the performance of CBIR. Data clustering is an unsupervised method for extraction hidden pattern from huge data sets. With large data sets, there is possibility of high dimensionality. Having both accuracy and efficiency for high dimensional data sets with enormous number of samples is a challenging arena. In this paper the clustering techniques are discussed and analysed. Also, we propose a method HDK that uses more than one clustering technique to improve the performance of CBIR.This method makes use of hierachical and divide and conquer KMeans clustering technique with equivalency and compatible relation concepts to improve the performance of the KMeans for using in high dimensional datasets. It also introduced the feature like color, texture and shape for accurate and effective retrieval system.
Algebraic Operations on PQ Trees and Modular Decomposition Trees
, 2005
"... Partitive set families are families of sets that can be quite large, but have a compact, recursive representation in the form of a tree. This tree is a common generalization of PQ trees, the modular decomposition of graphs, certain decompositions of boolean functions, and decompositions that arise o ..."
Abstract

Cited by 8 (1 self)
 Add to MetaCart
Partitive set families are families of sets that can be quite large, but have a compact, recursive representation in the form of a tree. This tree is a common generalization of PQ trees, the modular decomposition of graphs, certain decompositions of boolean functions, and decompositions that arise on a variety of other combinatorial structures. We describe natural operators on partitive set families, give algebraic identities for manipulating them, and describe efficient algorithms for evaluating them. We use these results to obtain new time bounds for finding the common intervals of a set of permutations, finding the modular decomposition of an edgecolored graphs (also known as a twostructure), finding the PQ tree of a matrix when a consecutiveones arrangement is given, and finding the modular decomposition of a permutation graph when its permutation realizer is given.
Homogeneity vs. adjacency: generalising some graph decomposition algorithms
 In 32nd International Workshop on GraphTheoretic Concepts in Computer Science (WG), volume 4271 of LNCS
, 2006
"... Abstract. In this paper, a new general decomposition theory inspired from modular graph decomposition is presented. Our main result shows that, within this general theory, most of the nice algorithmic tools developed for modular decomposition are still efficient. This theory not only unifies the usu ..."
Abstract

Cited by 6 (3 self)
 Add to MetaCart
(Show Context)
Abstract. In this paper, a new general decomposition theory inspired from modular graph decomposition is presented. Our main result shows that, within this general theory, most of the nice algorithmic tools developed for modular decomposition are still efficient. This theory not only unifies the usual modular decomposition generalisations such as modular decomposition of directed graphs and of 2structures, but also decomposition by star cutsets. 1
Practical and efficient circle graph recognition. arXiv:1104.3284
, 2011
"... ar ..."
(Show Context)
Dynamic distance hereditary graphs using split decomposition
 In Algorithms and Computation, 18th International Symposium (ISAAC
, 2007
"... Abstract. The problem of maintaining a representation of a dynamic graph as long as a certain property is satisfied has recently been considered for a number of properties. This paper presents an optimal algorithm for this problem on vertexdynamic connected distance hereditary graphs: both vertex i ..."
Abstract

Cited by 6 (4 self)
 Add to MetaCart
(Show Context)
Abstract. The problem of maintaining a representation of a dynamic graph as long as a certain property is satisfied has recently been considered for a number of properties. This paper presents an optimal algorithm for this problem on vertexdynamic connected distance hereditary graphs: both vertex insertion and deletion have complexity O(d), where d is the degree of the vertex involved in the modification. Our vertexdynamic algorithm is competitive with the existing linear time recognition algorithms of distance hereditary graphs, and is also simpler. Besides, we get a constant time edgedynamic recognition algorithm. To achieve this, we revisit the split decomposition by introducing graphlabelled trees. Doing so, we are also able to derive an intersection model for distance hereditary graphs, which answers an open problem. 1
Hierarchical clustering with CUDA/GPU
 in ISCA PDCCS
, 2009
"... Graphics processing units (GPUs) are powerful computational devices tailored towards the needs of the 3D gaming industry for highperformance, realtime graphics engines. Nvidia Corporation provides a programming language called CUDA for generalpurpose GPU programming. Hierarchical clustering i ..."
Abstract

Cited by 5 (0 self)
 Add to MetaCart
(Show Context)
Graphics processing units (GPUs) are powerful computational devices tailored towards the needs of the 3D gaming industry for highperformance, realtime graphics engines. Nvidia Corporation provides a programming language called CUDA for generalpurpose GPU programming. Hierarchical clustering is a common method used to determine clusters of similar data points in multidimensional spaces; if there are n data points, it can be computed in O(n2) to O(n2 log n) sequential time, depending on the distance metrics employed. The present work explores parallel computation of hierarchical clustering with CUDA/GPU, and obtains an overall speedup of up to 48 times over sequential computation with an Intel Pentium CPU. 1.