Accrue Software, Inc. Clustering is a division of data into groups of similar objects. Representing the data by fewer clusters necessarily loses certain fine details, but achieves simplification. It models data by its clusters. Data modeling puts clustering in a historical perspective rooted in mathematics, statistics, and numerical analysis. From a machine learning perspective clusters correspond to hidden patterns, the search for clusters is unsupervised learning, and the resulting system represents a data concept. From a practical perspective clustering plays an outstanding role in data mining applications such as scientific data exploration, information retrieval and text mining, spatial database applications, Web analysis, CRM, marketing, medical diagnostics, computational biology, and many others. Clustering is the subject of active research in several fields such as statistics, pattern recognition, and machine learning. This survey focuses on clustering in data mining. Data mining adds to clustering the complications of very large datasets with very many attributes of different types. This imposes unique

The detection of correlations between different features in a set of feature vectors is a very important data mining task because correlation indicates a dependency between the features or some association of cause and effect between them. This association can be arbitrarily complex, i.e. one or more features might be dependent from a combination of several other features. Well-known methods like the principal components analysis (PCA) can perfectly find correlations which are global, linear, not hidden in a set of noise vectors, and uniform, i.e. the same type of correlation is exhibited in all feature vectors. In many applications such as medical diagnosis, molecular biology, time sequences, or electronic commerce, however, correlations are not global since the dependency between features can be different in different subgroups of the set. In this paper, we propose a method called 4C (Computing Correlation Connected Clusters) to identify local subgroups of the data objects sharing a uniform but arbitrarily complex correlation. Our algorithm is based on a combination of PCA and density-based clustering (DBSCAN). Our method has a determinate result and is robust against noise. A broad comparative evaluation demonstrates the superior performance of 4C over competing methods such as DBSCAN, CLIQUE and ORCLUS.

Given an NN grid of squares, where each square has a count c i j and an underlying population p i j , our goal is to find the rectangular region with the highest density, and to calculate its significance by randomization. An arbitrary density function D, dependent on a region 's total count C and total population P, can be used. For example, if each count represents the number of disease cases occurring in that square, we can use Kulldorff's spatial scan statistic D K to find the most significant spatial disease cluster. A naive approach to finding the maximum density region requires O(N ) time, and is generally computationally infeasible. We present a multiresolution algorithm which partitions the grid into overlapping regions using a novel overlap-kd tree data structure, bounds the maximum score of subregions contained in each region, and prunes regions which cannot contain the maximum density region. For sufficiently dense regions, this method finds the maximum density region in ) time, in practice resulting in significant (20-2000x) speedups on both real and simulated datasets.

This material is posted here with permission from IEEE. Internal or personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution must be obtained from IEEE by writing to

Given an N N grid of squares, where each square s ij has a count c ij and an underlying population p ij , our goal is to nd the square region S with the highest density, and to calculate the signi cance of this region by Monte Carlo testing. Any density measure D, which depends on the total count and total population of the region, can be used. For example, if each count c ij represents the number of disease cases occurring in that square, we can use Kulldor's spatial scan statistic DK to nd the most signi cant spatial disease cluster. A naive approach to nding the region of maximum density would be to calculate the density measure for every square region: this requires O(RN ) calculations, where R is the number of Monte Carlo replications, and hence is generally computationally infeasible. We present a novel multi-resolution algorithm which partitions the grid into overlapping regions, bounds the maximum score of subregions contained in each region, and prunes regions which cannot contain the maximum density region. For suciently dense regions, this method nds the maximum density region in optimal O(RN ) time, and in practice it results in signi cant (10-200x) speedups as compared to the naive approach.

Assume a uniform, multidimensional grid of bivariate data, where each cell of the grid has a count ci and a baseline bi. Our goal is to find spatial regions (d-dimensional rectangles) where the ci are significantly higher than expected given bi. We focus on two applications: detection of clusters of disease cases from epidemiological data (emergency department visits, over-the-counter drug sales), and discovery of regions of increased brain activity corresponding to given cognitive tasks (from fMRI data). Each of these problems can be solved using a spatial scan statistic (Kulldorff, 1997), where we compute the maximum of a likelihood ratio statistic over all spatial regions, and find the significance of this region by randomization. However, computing the scan statistic for all spatial regions is generally computationally infeasible, so we introduce a novel fast spatial scan algorithm, generalizing the 2D scan algorithm of (Neill and Moore, 2004) to arbitrary dimensions. Our new multidimensional multiresolution algorithm allows us to find spatial clusters up to 1400x faster than the naive spatial scan, without any loss of accuracy. 1

The aim of this paper is to present a novel subspace clustering method named FINDIT. Clustering is the process of finding interesting patterns residing in the dataset by grouping similar data objects from dissimilar ones based on their dimensional values. Subspace clustering is a new area of clustering which achieves the clustering goal in high dimension by allowing clusters to be formed with their own correlated dimensions. In subspace clustering, selecting correct dimensions is very important because the distance between points is easily changed according to the selected dimensions. However, to select dimensions correctly is difficult, because data grouping and dimension selecting should be performed simultaneously. FINDIT determines the correlated dimensions for each cluster based on two key ideas: dimension-oriented distance measure which fully utilizes dimensional difference information, and dimension voting policy which determines important dimensions in a probabilistic way based on V nearest neighbors ’ information. Through various experiments on synthetic data, FINDIT is shown to be very successful in the high dimensional clustering problem. FINDIT satisfies most requirements for good clustering methods such as accuracy of results, robustness to the noise and the cluster density, and scalability to the dataset size and the dimensionality. Moreover, it is gracefully scalable to full dimension without any modification to algorithm.

Abstract—We present the first framework for comparing subspace clusterings. We propose several distance measures for subspace clusterings, including generalizations of well-known distance measures for ordinary clusterings. We describe a set of important properties for any measure for comparing subspace clusterings and give a systematic comparison of our proposed measures in terms of these properties. We validate the usefulness of our subspace clustering distance measures by comparing clusterings produced by the algorithms FastDOC, HARP, PROCLUS, ORCLUS, and SSPC. We show that our distance measures can be also used to compare partial clusterings, overlapping clusterings, and patterns in binary data matrices. Index Terms—Subspace clustering, projected clustering, distance, feature selection, cluster validation.

Clustering is one of the central data mining problems and numerous approaches have been proposed in this field. However, few of these methods focus on categorical data. The categorical techniques that do exist have significant shortcomings in terms of performance, the clusters they detect, and their ability to locate clusters in subspaces.

Application domains such as life sciences, e.g. molecular biology produce a tremendous amount of data which can no longer be managed without the help of e#cient and e#ective data mining methods. One of the primary data mining tasks is clustering. However, traditional clustering algorithms often fail to detect meaningful clusters because of the high dimensional, inherently sparse feature space of most real-world data sets. Nevertheless, the data sets often contain clusters hidden in various subspaces of the original feature space. We present a pre-processing step for traditional clustering algorithms, which detects all interesting subspaces of high-dimensional data containing clusters. For this purpose, we define a quality criterion for the interestingness of a subspace and propose an e#cient algorithm called RIS (Ranking I nteresting Subspaces) to examine all such subspaces. A broad evaluation based on synthetic and real-world data sets empirically shows that RIS is suitable to find all relevant subspaces in large, high dimensional, sparse data and to rank them accordingly.