M. Ester, H. Kriegel, and X. Xu. A Database Interface for Clustering in Large Spatial Databases. In Proceedings of the First International Conference on Knowledge Discovery and Data Mining, 1995.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....and have a wide range of applications, such as data compression, information retrieval, pattern recognition, trend analysis, customer segmentation and classi cation. The problem of clustering has been studied extensively in the database [Zhang et al. 1996; Guha et al. 1998; Ng and Han, 1994; Ester et al. 1995a; Ester et al. 1995c; Ester et al. 1995b; Li et al. 2001] statistics [Brito et al. 1997; Berger and Rigoutsos, 1991; Duda and Hart, 1973; Dubes and Jain, 1980; Lee, 1981; Murtagh, 1983] and machine learning communities [Cheeseman et al. 1988; Fisher, 1987; Fisher, 1995; Lebowitz, 1987; Liu ....

....of applications, such as data compression, information retrieval, pattern recognition, trend analysis, customer segmentation and classi cation. The problem of clustering has been studied extensively in the database [Zhang et al. 1996; Guha et al. 1998; Ng and Han, 1994; Ester et al. 1995a; Ester et al. 1995c; Ester et al. 1995b; Li et al. 2001] statistics [Brito et al. 1997; Berger and Rigoutsos, 1991; Duda and Hart, 1973; Dubes and Jain, 1980; Lee, 1981; Murtagh, 1983] and machine learning communities [Cheeseman et al. 1988; Fisher, 1987; Fisher, 1995; Lebowitz, 1987; Liu et al. 2000] with ....

[Article contains additional citation context not shown here]

Martin Ester, Hans-Peter Kriegel, and Xiaowei Xu, \A Database Interface for Clustering in Large Spatial Databases," In Proc. of 1st Int'l Conf. on KDD, 1995. 17

....an item for each categorical value. However, our method is specifically applicable to the case of discovering useful associations in market basket data, as opposed to finding well partitioned clusters in categorical data. The problem of clustering has been widely studied in the literature [5, 6, 9, 10, 11, 13, 14, 15, 16, 18, 22, 25, 26]. In recent years, the importance of clustering categorical data has received considerable attention from researchers [13, 14, 16, 17] In [17] a clustering technique is proposed in which clusters of items are used in order to cluster the points. The merit in this approach is that it recognized ....

M. Ester, H.-P. Kriegel, X. Xu. "A Database Interface for Clustering in Large Spatial Databases", Proceedings of the Knowledge Discovery and Data Mining Conference, 1995.

.... K Means algorithm [12, 22] Different improvements to alleviate effects of initialization on results of KMeans are proposed [4, 34] While we are not concerned here with scalability issues, we note that classic K Means has numerous extensions to scalable unsupervised learning in databases CLARANS [14] and BIRCH [35] IR algorithm is an iterative optimization of objective function (equal to a reduction in mutual information) It can be viewed in a context of general EM framework [9, 28, 30] Two specific examples of particular algorithms are AutoClass [6] and MCLUST [18] New approaches are ....

....10. Screen Shots File View Tools Help IDIxI Current View: Analysis Normal Data Set Cols: 203 Top Group Cols: 22 [20] harcourtonline. COM [3] isseI.UK [3] itmcenter. COM [1] about. COM [3] google. COM [12] excite. COM [5] n a. NUM [1] digitalmass.COM [14] bolt. COM [8] collegeclub. COM [8] autozone. COM [68] searchenginewatch. COM [6] microsoit. COM [1 ] webtop. COM [5] scripps.COM [1] rsvp0.NET Figure 1: Original 197 rows X representing different referrers and 203 columns Y representing important pages are clustered into 26 row ....

Ester, M., Kriegel, H-P., Xu, X., A Database interface for Clustering in Large Spatial Databases, Proceedings Of the 1 st Intl. Conf. KDD, Montreal, Canada, 1995.

....a local minimum is found, and the algorithm restarts until humlocal local minima are found (value humlocal=2 is recommended) The best node (set of reedolds) is returned for the formation of a resulting partition. The complexity of CLARANS is O(N 2) in terms of number of points. Ester et al. [EKX95] extended CLARANS to spatial VLDB. They used R trees [BKS90] to relax the original requirement that all the data resides in core memory, which allowed focusing exploration on the relevant part of the database that resides at a branch of the whole data tree. The k means algorithm [Har75] ....

....this kind of task because centroids become centers of weights instead of means. Sometimes this described practice is called case scaling. Some algorithms depend on the effectiveness of data access. To facilitate this process data indices are constructed. Examples include the extension of CLARANS [EKX95] and 37 the algorithm DBSCAN [EKSX96] Index structures used for spatial data, include KDtrees [FBF77] R trees [Gut84] R trees [KSSB90] A blend of attribute transformations (DFT, Polynomials) and indexing technique is presented in [KCPM01 ] Other indices and numerous generalizations exist ....

Ester, M., Kriegel, H-P., and Xu, X. A database interface for clustering in large spatial databases. KDD-95, Montreal, Canada, 1995.

.... large patterns. By the phrase large patterns , we mean characteristics of the input data that are exhibited by a (typically userdefined) significant portion of the data. Examples of these large patterns include association rules[AMS 95] classification[RS98] and clustering[ZRL96, NH94, EKX95, GRS98] In this paper, we focus on the converse problem of finding small patterns or outliers. An outlier in a set of data is an observation or a point that is considerably dissimilar or inconsistent with the remainder of the data. From the above description of outliers, it may seem that ....

....the problem with these techniques is that they rely on computing ffi dimensional convex hulls which is inherently an exponential process with a lower bound of Omega Gamma dffi=2e ) This makes these techniques infeasible for dimensions 2. Clustering algorithms like CLARANS [NH94] DBSCAN [EKX95] BIRCH [ZRL96] and CURE [GRS98] consider outliers, but only to the point of ensuring that they do not interfere with the clustering process. Further, the definition of outliers used is in a sense subjective and related to the clusters that are detected by these algorithms. This is in contrast to ....

Martin Ester, Hans-Peter Kriegel, and Xiaowei Xu. A database interface for clustering in large spatial databases. In Int'l Conference on Knowledge Discovery in Databases and Data Mining (KDD-95), Montreal, Canada, August 1995.

....including engineering, business and social science, and have a wide range of applications, such as data compression, information retrieval, pattern recognition, trend analysis, customer segmentation and classi cation. The problem of clustering has been studied extensively in the database [43, 18, 33, 12, 14, 13], statistics [5, 4, 9, 8, 34, 32] and machine learning communities [7, 15, 16, 29, 30] with di erent approaches and di erent focuses. The clustering problem can be described as follows: let V be a set of n multi dimensional data points, we want to nd a partition of V into clusters such that the ....

Martin Ester, Hans-Peter Kriegel, and Xiaowei Xu. A database interface for clustering in large spatial databases. In Proc. of 1st Int'l Conf. on KDD, 1995.

....designed for arbitrary insertions and deletions of transactions and hence do not exploit systematic block evolution. Moreover, they do not consider and cannot maintain models for the most recent window option with respect to an arbitrary block selection sequence. Ester et al. 7] extended DBScan [8] to develop a scalable incremental clustering algorithm. In prior work, we developed a scalable incremental algorithm for maintaining decision tree classifiers [11] Utgoff et al. 17] developed ID5, an incremental version of ID3, which assumes that the entire dataset fits in main memory and hence ....

M. Ester, H.-P. Kriegel, and X. Xu. A database interface for clustering in large spatial databases. In Proc. of the 1st Int'l Conference on Knowledge Discovery in Databases and Data Mining, Montreal, Canada, August 1995.

.... large patterns. By the phrase large patterns , we mean characteristics of the input data that are exhibited by a (typically user defined) significant portion of the data. Examples of these large patterns include association rules[AMS 95] classification[RS98] and clustering[ZRL96, NH94, EKX95, GRS98] In this paper, we focus on the converse problem of finding small patterns or outliers. An outlier in a set of data is an observation or a point that is considerably dissimilar or inconsistent with the remainder of the data. From the The work was done while the author was with Bell ....

....on real life and synthetic databases. Section 7 concludes the paper. The work reported in this paper has been done in the context of the Serendip data mining project at Bell Laboratories (www.bell labs.com projects serendip) 2 Related Work Clustering algorithms like CLARANS [NH94] DBSCAN [EKX95] BIRCH [ZRL96] and CURE [GRS98] consider outliers, but only to the point of ensuring that they do not interfere with the clustering process. Further, the definition of outliers used is in a sense subjective and related to the clusters that are detected by these algorithms. This is in contrast to ....

Martin Ester, Hans-Peter Kriegel, and Xiaowei Xu. A database interface for clustering in large spatial databases. In Int'l Conference on Knowledge Discovery in Databases and Data Mining (KDD-95), Montreal, Canada, August 1995.

....The clustering problem has been discussed extensively in the database literature as a tool for similarity search, customer segmentation, pattern recognition, trend analysis and classification. Various methods have been studied in considerable detail by both the statistics and database communities [3, 4, 7, 8, 9, 13, 21, 26]. Detailed surveys on clustering methods can be found in [6, 17, 18, 20, 25] The problem of clustering data points is defined as follows: Given a set of points in multidimensional space, find a partition of the points into clusters so that the points within each cluster are close to one another. ....

M. Ester, H.-P. Kriegel, X. Xu. A Database Interface for Clustering in Large Spatial Databases. Proceedings of the first International Conference on Knowledge Discovery and Data Mining, 1995.

....of the current node is set to be the current node if it results in better clustering. It is experimentally shown that CLARANS outperforms the traditional k medoid algorithms. CLARANS, however, suffers from the same drawbacks as the other partitional clustering algorithms described earlier. In [EKX95], the authors use the R tree[SRF87, BKSS90, Sam89] to improve the I O efficiency of CLARANS on large databases by 1) drawing samples from leaf pages to reduce the number of data points (since data points are packed in leaf nodes based on spatial locality, a sample point in the leaf page can be a ....

Martin Ester, Hans-Peter Kriegel, and Xiaowei Xu. A database interface for clustering in large spatial databases. In Int'l Conference on Knowledge Discovery in Databases and Data Mining (KDD-95), Montreal, Canada, August 1995.

....of the data object. For clusters with arbitrary shapes, these algorithms connect regions with sufficiently high densities into clusters. In this section, we will discuss three major algorithms under this category. DBSCAN DBSCAN (Density Based Spatial Clustering of Applications with Noise) [15] defines a cluster as a maximal set of density connected data objects. The key idea behind this algorithm is that for each data object in a cluster, the data object s neighbourhood of a given radius (ffl) has to contain at least a minimum number of data objects (MinP ts) In other words, the ....

....closer to o random than to any other centres, o i where 1 i k, might potentially affect the test result. Hence while testing the quality of o random , we only have to focus on data objects that fulfill this criteria. Those data objects can be located by making a boundary query using the R Tree [15]. In an obstacle planar space, it is difficult to find spatial proximity with R Tree. The bounded query result does not guarantee that the data points are closer to o random than o i due to the obstruction. We alter the technique by sacrificing some memory space for speed and efficiency. In line ....

M. Ester, H.-P. Kriegel, and Xu X. A database interface for clustering in large spatial databases. In Proceedings of the 1st International Conference on Knowledge Discovery and Data Mining, pages 94--99. Montreal, Canada, 1995.

....been developed. Since these methods have been designed speci cally for processing very large data sets, they may be applicable with some modi cations to Web based information retrieval systems. Examples of some of these techniques are given in [Agrawal et al. 1998] Dhillon 1998] Dhillon 1999] [Ester et al. 1995a] Ester et al. 1995b] Ester et al. 1995c] Fisher 1995] Guha et al. 1998] Ng, Han 1994] Zhang et al. 1996] For very large databases, apropriate parallel algorithms can speed up computations [Omiecinski, Scheuermann 1990] Finally, we note that Clustering is just one of several ways ....

....these methods have been designed speci cally for processing very large data sets, they may be applicable with some modi cations to Web based information retrieval systems. Examples of some of these techniques are given in [Agrawal et al. 1998] Dhillon 1998] Dhillon 1999] Ester et al. 1995a] [Ester et al. 1995b] Ester et al. 1995c] Fisher 1995] Guha et al. 1998] Ng, Han 1994] Zhang et al. 1996] For very large databases, apropriate parallel algorithms can speed up computations [Omiecinski, Scheuermann 1990] Finally, we note that Clustering is just one of several ways of organizing ....

[Article contains additional citation context not shown here]

Ester, M., Kriegel, H.-S., Xu, X., \A database interface for clustering in large spatial databases", Proc. First Int'l. Conference on Knowledge Discovery in Data Bases and Data Mining, AAAI Press, Menlo Park, CA (1995).

No context found.

M. Ester, H.-P. Kriegel, and X. Xu, "A Database Interface for Clustering in Large Spatial Databases", In Proc. 1 st Int. Conf. on Knowledge Discovery and Data Mining (KDD95), Montreal, Canada, 1995, pp. 94-99.

No context found.

M. Ester, H. Kriegel, and X. Xu. A Database Interface for Clustering in Large Spatial Databases. In Proceedings of the First International Conference on Knowledge Discovery and Data Mining, 1995.

No context found.

Martin Ester, Hans-Peter Kriegel, and Xiaowei Xu, \A Database Interface for Clustering in Large Spatial Databases," In Proc. of 1st Int'l Conf. on KDD, 1995. 17

No context found.

Martin Ester, Hans-Peter Kriegel, and Xiaowei Xu. A database interface for clustering in large spatial databases. In Int. Conference on Knowledge Discovery in Databases and Data Mining (KDD-95), Montreal, Cananda, August 1995.

No context found.

M. Ester, H. P. Kriegel, and X. Xu. A database interface for clustering in large spatial databases. In Proc. KDD, 1995.

No context found.

Ester M., Kriegel H-P., Xu X.: A Database Interface for Clustering in Large Spatial Databases. Proc. of the 1st Int'l Conference on Knowledge Discovery and Data Mining (KDD), Montreal, Canada (1995)

No context found.

Martin Ester, Hans-Peter Kriegel, and Xiaowei Xu. A database interface for clustering in large spatial databases. In Int'l Conference on Knowledge Discovery in Databases and Data Mining (KDD-95), Montreal, Cananda, August 1995.

No context found.

Martin Ester, Hans-Peter Kriegel, and Xiaowei Xu. A database interface for clustering in large spatial databases. In Proc. of the 1st Int'l Conference on Knowledge Discovery in Databases and Data Mining, Montreal, Canada, August 1995.

No context found.

Martin Ester, Hans-Peter Kriegel, and Xiaowei Xu. A database interface for clustering in large spatial databases. In Proc. of the 1st Int'l Conference on Knowledge Discovery in Databases and Data Mining, Montreal, Canada, August 1995.

No context found.

M. Ester, H. P. Kriegel, and X. Xu. A database interface for clustering in large spatial databases. In Proc. KDD, 1995.

No context found.

M. Ester, H.-P. Kriegel, and X. Xu. A database interface for clustering in large spatial databases. KDD, 1995.

No context found.

M. Ester, H-P. Kriegel, X. Xu, A Database Interface for Clustering in Large Spatial Databases, Proc. 1st Int. Conf. on Knowledge Discovery and Data Mining, 1995.

No context found.

Ester M., Kriegel H.-P., Xu X.: `A Database Interface for Clustering in Large Spatial Databases', Proc. 1st Int. Conf. on Knowledge Discovery and Datamining (KDD95) , Montreal, Canada, 1995.

First 50 documents

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC