### Table 2: Clustering Algorithm

"... In PAGE 7: ...he correlation information (i.e. principal components) to obtain the correlated clusters. The clustering algorithm is shown in Table2 . It takes a set of points A and a set of clusters S as input.... ..."

### Table 5 Clustering algorithms

2006

"... In PAGE 8: ... The Gaussians are spherical, but they may have different volumes. Table5 presents a list of the clustering algorithms used in the paper. A more complete list is used for the companion web page.... ..."

### Table 2: Clustering Algorithm

2000

Cited by 91

### Table 2: Clustering Algorithm

2000

Cited by 91

### Table 3. SOS clustering algorithm

"... In PAGE 7: ...2.2 Self Organizing Sensor (SOS)Clustering Algorithm Table3 illustrates the pseudo code of the SOS clustering algorithm and it consists of 5 parts. Table 3.... ..."

Cited by 1

### Table 3: Iterative Clustering Algorithm

2000

"... In PAGE 5: ... To determine the number of dimensions to be retained for each cluster , we first determine, for each point , the best cluster, if one exists, for placing . Let denote the the least di- mensionality needed for the cluster to represent with 4For subsequent invocations of FindClusters procedure during the iterative algorithm (Step 2 in Table3 ), there may exist already completed clusters (does not exist during the initial invocation). Hence must also be sufficiently far from all complete clusters formed so far i.... ..."

Cited by 91

### Table 3: Iterative Clustering Algorithm

2000

"... In PAGE 7: ... 4 This technique, proposed by Gonzalez [22], is guaranteed to return a piercing if no outliers are present. To avoid scanning though the whole database 4For subsequent invocations of FindClusters procedure during the iterative algorithm (Step 2 in Table3 ), there may exist already completed clusters (does not exist during the initial invocation). Hence P must also be sufficiently far from all complete clusters formed so far i.... ..."

Cited by 91

### Table 1: Clustering algorithms and their labels.

"... In PAGE 8: ... We also tested the performance of a baseline \no- clustering quot; option, for which we simply used the HY(n,u) algorithm with no clustering extensions. Table1 summa- rizes the clustering algorithms and the labels that we will use to identify them when describing the results. We con- sidered the possibility of using an existing OODB bench- mark such as OO7 [CDN93] for these experiments; however, benchmarks like OO7 have placed too little emphasis on clustering-related operations to be e ective for this purpose [CDK+94], and the planned extensions to OO7 in this area were never completed.... ..."

### Table 3: Iterative Clustering Algorithm

"... In PAGE 7: ... 4 This technique, proposed by Gonzalez [22], is guaranteed to return a piercing if no outliers are present. To avoid scanning though the whole database 4For subsequent invocations of FindClusters procedure during the iterative algorithm (Step 2 in Table3 ), there may exist already completed clusters (does not exist during the initial invocation). Hence P must also be sufficiently far from all complete clusters formed so far i.... ..."

### Table 1. Complexity of Clustering Algorithms

"... In PAGE 35: ... It is possible to compute the entries of this matrix based on need instead of storing them (but this would increase the algorithm apos;s time complexity [7]). Table1 lists the time and space complexities of several well-known algorithms. Here, n is the number of patterns to be clustered, k is the number of clusters, and l is the number of iterations.... ..."