Data Clustering

Clustering is the unsupervised classification of patterns (observations, data items, or feature vectors) into groups (clusters). The clustering problem has been addressed in many contexts and by researchers in many disciplines; this reflects its broad appeal and usefulness as one of the steps in exploratory data analysis. However, clustering is a difficult problem combinatorially, and differences in assumptions and contexts in different communities has made the transfer of useful generic concepts and methodologies slow to occur. This paper presents an overview of pattern clustering methods from a statistical pattern recognition perspective, with a goal of providing useful advice and references to fundamental concepts accessible to the broad community of clustering practitioners. We present a taxonomy of clustering techniques, and identify cross-cutting themes and recent advances. We also describe some important applications of clustering algorithms such as image segmentation, object recognition, and information retrieval.

Finding useful patterns in large datasets has attracted considerable interest recently, and one of the most widely st,udied problems in this area is the identification of clusters, or deusel y populated regions, in a multi-dir nensi onal clataset. Prior work does not adequately address the problem of large datasets and minimization of 1/0 costs.

Clustering, in data mining, is useful to discover distribution patterns in the underlying data. Clustering algorithms usually employ a distance metric based (e.g., euclidean) similarity measure in order to partition the database such that data points in the same partition are more similar than points in different partitions. In this paper, we study clustering algorithms for data with boolean and categorical attributes. We show that traditional clustering algorithms that use distances between points for clustering are not appropriate for boolean and categorical attributes. Instead, we propose a novel concept of links to measure the similarity/proximity between a pair of data points. We develop a robust hierarchical clustering algorithm ROCK that employs links and not distances when merging clusters. Our methods naturally extend to non-metric similarity measures that are relevant in situations where a domain expert/similarity table is the only source of knowledge. In addition to presenting detailed complexity results for ROCK, we also conduct an experimental study with real-life as well as synthetic data sets to demonstrate the effectiveness of our techniques. For data with categorical attributes, our findings indicate that ROCK not only generates better quality clusters than traditional algorithms, but it also exhibits good scalability properties.

Hierarchical clustering is a widely used method for detecting clusters in genomic data. Clusters are defined by cutting branches off the dendrogram. A common but inflexible method uses a constant height cutoff value; this method exhibits suboptimal performance on complicated dendrograms. We present the Dynamic Tree Cut R package that implements novel dynamic branch cutting methods for detecting clusters in a dendrogram depending on their shape. Compared to the constant height cutoff method, our techniques offer the following advantages: (1) they are capable of identifying nested clusters; (2) they are flexible-cluster shape parameters can be tuned to suit the application at hand; (3) they are suitable for automation; and (4) they can optionally combine the advantages of hierarchical clustering and partitioning around medoids, giving better detection of outliers. We illustrate the use of these methods by applying them to protein-protein interaction network data and to a simulated gene expression data set. Availability: The Dynamic Tree Cut method is implemented in an R

| Discovering association rules is one of the most important task in data mining. Many e cient algorithms have been proposed in the literature. The most noticeable are Apriori, Mannila's algorithm, Partition, Sampling and DIC, that are all based on the Apriori mining method: pruning the subset lattice (itemset lattice). In this paper we propose an e cient algorithm, called Close, based on a new mining method: pruning the closed set lattice (closed itemset lattice). This lattice, which is a sub-order of the subset lattice, is closely related to Wille's concept lattice in formal concept analysis. Experiments comparing Close to an optimized version of Apriori showed that Close is very e cient for mining dense and/or correlated data such as census style data, and performs reasonably well for market basket style data.

Traditional problem determination techniques rely on static dependency models that are difficult to generate accurately in today's large, distributed, and dynamic application environments such as e-commerce systems. In this paper, we present a dynamic analysis methodology that automates problem determination in these environments by 1) coarse-grained tagging of numerous real client requests as they travel through the system and 2) using data mining techniques to correlate the believed failures and successes of these requests to determine which components are most likely to be at fault. To validate our methodology, we have implemented Pinpoint, a framework for root cause analysis on the J2EE platform that requires no knowledge of the application components. Pinpoint consists of three parts: a communications layer that traces client requests, a failure detector that uses traffic-sniffing and middleware instrumentation, and a data analysis engine. We evaluate Pinpoint by injecting faults into various application components and show that Pinpoint identifies the faulty components with high accuracy and produces few false-positives.

Data clustering is an important technique for exploratory data analysis, and has been studied for several years. It has been shown to be useful in many practical domains such as data classification and image processing. Recently, there has been a growing emphasis on exploratory analysis of very large datasets to discover useful patterns and/or correlations among attributes. This is called data mining, and data clustering is regarded as a particular branch. However existing data clustering methods do not adequately address the problem of processing large datasets with a limited amount of resources (e.g., memory and cpu cycles). So as the dataset size increases, they do not scale up well in terms of memory requirement, running time, and result quality.

The graph construction procedure essentially determines the potentials of those graph-oriented learning algorithms for image analysis. In this paper, we propose a process to build the so-called directed 1 -graph, in which the vertices involve all the samples and the ingoing edge weights to each vertex describe its 1 -norm driven reconstruction from the remaining samples and the noise. Then, a series of new algorithms for various machine learning tasks, e.g., data clustering, subspace learning, and semisupervised learning, are derived upon the 1 -graphs. Compared with the conventional -nearest-neighbor graph and -ball graph, the 1 -graph possesses the advantages: 1) greater robustness to data noise, 2) automatic sparsity, and 3) adaptive neighborhood for individual datum. Extensive experiments on three real-world datasets show the consistent superiority of 1 -graph over those classic graphs in data clustering, subspace learning, and semi-supervised learning tasks.

Partitioning a data set and extracting hidden structure from the data arises in different application areas of pattern recognition, speech and image processing. Pairwise data clustering is a combinatorial optimization method for data grouping which extracts hidden structure from proximity data. We describe a deterministic annealing approach to pairwise clustering which shares the robustness properties of maximum entropy inference. The resulting Gibbs probability distributions are estimated by mean-field approximation. A new structure-preserving algorithm to cluster dissimilarity data and to simultaneously embed these data in a Euclidian vector space is discussed which can be used for dimensionality reduction and data visualization. The suggested embedding algorithm which outperforms conventional approaches has been implemented to analyze dissimilarity data from protein analysis and from linguistics. The algorithm for pairwise data clustering is used to segment textured images.

This paper describes the application of adaptive neuro-fuzzy inference system (ANFIS) model for classification of electroencephalogram (EEG) signals. Decision making was performed in two stages: feature extraction using the wavelet transform (WT) and the ANFIS trained with the backpropagation gradient descent method in combination with the least squares method. Five types of EEG signals were used as input patterns of the five ANFIS classifiers. To improve diagnostic accuracy, the sixth ANFIS classifier (combining ANFIS) was trained using the outputs of the five ANFIS classifiers as input data. The proposed ANFIS model combined the neural network adaptive capabilities and the fuzzy logic qualitative approach. Some conclusions concerning the saliency of features on classification of the EEG signals were obtained through analysis of the ANFIS. The performance of the ANFIS model was evaluated in terms of training performance and classification accuracies and the results confirmed that the proposed ANFIS model has potential in classifying the EEG signals.

We explore the idea of evidence accumulation for combining the results of multiple clusterings. Initially, n d−dimensional data is decomposed into a large number of compact clusters; the K-means algorithm performs this decomposition, with several clusterings obtained by N random initializations of the K-means. Taking the cooccurrences of pairs of patterns in the same cluster as votes for their association, the data partitions are mapped into a co-association matrix of patterns. This n × n matrix represents a new similarity measure between patterns. The final clusters are obtained by applying a MST-based clustering algorithm on this matrix. Results on both synthetic and real data show the ability of the method to identify arbitrary shaped clusters in multidimensional data.

This paper presents a computational paradigm called Data Driven Markov Chain Monte Carlo (DDMCMC) for image segmentation in the Bayesian statistical framework. The paper contributes to image seymentation in three aspects. Firstly, it designs effective and well balanced Markov Chain dynamics to explore the solution space and makes the split and merge process reversible at a middle level vision formulation. Thus at achieves globally optimal solution independent of initial segmentatzons. Secondly, instead of computing a single maximum a posteriori solution, it proposes a mathematical principle f o r computing multiple distinct solutions to incorporates intrinsic ambiguities an image segmentation. A k-adventurers algorithm is proposed f o r extracting distinct multiple solutions f r o m the Markov chain sequence. Thirdly, it utilizes datadriven (bottom-up) techniques, such as clustering and edge detection, to compute importance proposal probabilities, which effectively drive the Markov chain dynamics and achieve tremendous speedup in comparison to traditional jump-diffusion method[J]. Thus DDM-

Data clustering describes a set of frequently employed techniques in exploratory data analysis to extract "natural" group structure in data. Such groupings need to be validated to separate the signal in the data from spurious structure. In this context, nding an appropriate number of clusters is a particularly important model selection question. We introduce a measure of cluster stability to assess the validity of a cluster model. This stability measure quanti es the reproducibility of clustering solutions on a second sample, and it can be interpreted as a classi cation risk with regard to class labels produced by a clustering algorithm. The preferred number of clusters is determined by minimizing this classi cation risk as a function of the number of clusters. Convincing results are achieved on simulated as well as gene expression data sets. Comparisons to other methods demonstrate the competitive performance of our method and its suitability as a general validation tool for clustering solutions in realworld problems.

This paper tries to reconcile two sets of apparently contradictory results. One is the positive link, postulated in literature, between place attachment and civic activity, the other is the sociological claim that there is a negative relationship between place attachment and a person's social and cultural status (cultural capital), which in turn is known to be the strongest positive predictor of civic involvement.

Background: Data clustering analysis has been extensively applied to extract information from gene expression profiles obtained with DNA microarrays. To this aim, existing clustering approaches, mainly developed in computer science, have been adapted to microarray data analysis. However, previous studies revealed that microarray datasets have very diverse structures, some of which may not be correctly captured by current clustering methods. We therefore approached the problem from a new starting point, and developed a clustering algorithm designed to capture dataset-specific structures at the beginning of the process. Results: The clustering algorithm is named Fuzzy clustering by Local Approximation of MEmbership (FLAME). Distinctive elements of FLAME are: (i) definition of the neighborhood of each object (gene or sample) and identification of objects with "archetypal" features named Cluster Supporting Objects, around which to construct the clusters; (ii) assignment to each object of a fuzzy membership vector approximated from the memberships of its neighboring objects, by an iterative converging process in which membership spreads from the Cluster Supporting Objects through their neighbors. Comparative analysis with K-means, hierarchical, fuzzy C-means and fuzzy self-organizing maps (SOM) showed that data partitions generated by FLAME are not superimposable to those of other methods and, although different types of datasets are better partitioned by different algorithms, FLAME displays the best overall performance. FLAME is implemented, together with all the above-mentioned algorithms, in a C++ software with graphical interface for Linux and Windows, capable of handling very large datasets, named Gene Expression Data Analysis Studio (GEDAS), freely available under GNU General Public License. Conclusion: The FLAME algorithm has intrinsic advantages, such as the ability to capture nonlinear relationships and non-globular clusters, the automated definition of the number of clusters, and the identification of cluster outliers, i.e. genes that are not assigned to any cluster. As a result, clusters are more internally homogeneous and more diverse from each other, and provide better partitioning of biological functions. The clustering algorithm can be easily extended to applications different from gene expression analysis.

This paper explores basic aspects of the immune system and proposes a novel immune network model with the main goals of clustering and filtering unlabeled numerical data sets. It is not our concern to reproduce with confidence any immune phenomenon, but to show that immune concepts can be used to develop powerful computational tools for data processing. As important results of our model, the network evolved will be capable of reducing redundancy, describing data structure, including the shape of the clusters. The network will be implemented in association with a statistical inference technique, and its performance will be illustrated using two benchmark problems. The paper is concluded with a trade-off between the proposed network and artificial neural networks used to perform unsupervised learning.

Despite its benefit in a wide range of applications, data mining techniques also have raised a number of ethical issues. Some such issues include those of privacy, data security, intellectual property rights, and many others. In this paper, we address the privacy problem against unauthorized secondary use of information. To do so, we introduce a family of geometric data transformation methods (GDTMs) which ensure that the mining process will not violate privacy up to a certain degree of security. We focus primarily on privacy preserving data clustering, notably on partition-based and hierarchical methods. Our proposed methods distort only confidential numerical attributes to meet privacy requirements, while preserving general features for clustering analysis. Our experiments demonstrate that our methods are effective and provide acceptable values in practice for balancing privacy and accuracy. We report the main results of our performance evaluation and discuss some open research issues.

The Fuping Complex, located within the central zone of the North China Craton, is composed of amphibolite to granulite facies orthogneisses, interleaved with minor supracrustal rocks at similar metamorphic grade. The oldest components recognised are hornblende gneiss enclaves within the predominant biotite orthogneiss which have a SHRIMP U-Pb zircon age of 2708 98 Ma. We consider these enclaves to represent fragments of 2.7 Ga continental materials incorporated in the biotite gneiss. The biotite gneiss has a SHRIMP U -Pb zircon age of 2513 912 Ma, interpreted to be time of magmatic crystallisation of the igneous precursor, based on the igneous characteristics of the zircons. This indicates a major magmatic episode at 2.52 Ga in the Fuping Complex, identical to the age of felsic volcanism within the low-grade Wutai Complex which crops out immediately to the west. A gneissic granite that intrudes the biotite gneisses has a poorly-defined 207 Pb/ 206 Pb age of 2045 9 64 Ma. This is within error of the age of 2097 9 46 Ma obtained from a fine-grained gneiss interlayered with amphibolite of the Wanzi Supracrustal Suite (WSS), interpreted to be volcanic in origin. Zircons from both these samples have strong oscillatory zoning and provide the first indication of a Palaeoproterozoic magmatic event in the area, again similar in age to magmatic events recently recognised in the adjacent Wutai Complex. These data indicate a comparable history for the Fuping and Wutai Complexes and support geochemical evidence that they had a common origin and formed part of a Late Archaean arc, affected by later Palaeoproterozoic re-activation. Low-uranium zircons without oscillatory zoning, separated from a sample of biotite gneiss, yield data clustered at 1817 926 Ma, which is interpreted to reflect a period of new zircon growth during a major metamorphic event. These data support the recently-proposed tectonic model that amalgamation of the North China Craton occurred due to collision of the Eastern and Western Blocks along the central zone at 1.8 Ga ago.

We have developed a simple and expandable procedure for classification and validation of extracellular data based on a probabilistic model of data generation. This approach relies on an empirical characterization of the recording noise. We first use this noise characterization to optimize the clustering of recorded events into putative neurons. As a second step, we use the noise model again to assess the quality of each cluster by comparing the within-cluster variability to that of the noise. This second step can be performed independently of the clustering algorithm used, and it provides the user with quantitative as well as visual tests of the quality of the classification. #

Until now, neighbor-embedding-based (NE) algorithms for super-resolution (SR) have carried out two independent processes to synthesize high-resolution (HR) image patches. In the first process, neighbor search is performed using the Euclidean distance metric, and in the second process, the optimal weights are determined by solving a constrained least squares problem. However, the separate processes are not optimal. In this paper, we propose a sparse neighbor selection scheme for SR reconstruction. We first predetermine a larger number of neighbors as potential candidates and develop an extended Robust-SL0 algorithm to simultaneously find the neighbors and to solve the reconstruction weights. Recognizing that the -nearest neighbor ( -NN) for reconstruction should have similar local geometric structures based on clustering, we employ a local statistical feature, namely histograms of oriented gradients (HoG) of low-resolution (LR) image patches, to perform such clustering. By conveying local structural information of HoG in the synthesis stage, the -NN of each LR input patch is adaptively chosen from their associated subset, which significantly improves the speed of synthesizing the HR image while preserving the quality of reconstruction. Experimental results suggest that the proposed method can achieve competitive SR quality compared with other state-of-the-art baselines.

Veins formed in a variety of rock types and tectonic environments were measured at seven field locations to determine a general scaling relationship between the length and opening displacement (aperture). For these naturally formed extension fractures displacement-length profiles for both single-and multiplesegment veins display centrally located maxima and gently tapered displacement gradients at the tips. This geometry is consistent with elastic-plastic crack growth models, which also predict linear scaling between length and displacement. The geometry is used to estimate stresses of formation for the veins. Length-maximum displacement plots for single-segment fractures show data clustered along linear trends. Similar plots for multiple-segment fractures show greater scatter and are marginally better fit by a square root function than by a line. The geometries of multiple-segment veins are consistent with elastic models for crack-tip stress interactions. Aspect ratios for single-segment fracture sets vary between 1 x 1~ 3 and 8 x 10 -`3 , consistent with tensional fracture strengths roughly an order of magnitude lower than compressive shear strengths implied by shear displacement-length ratios recorded for faults, which are about 10 2.

Group work is widespread in education. The growing use of online tools supporting group work generates huge amounts of data. We aim to exploit this data to support mirroring: presenting useful high-level views of information about the group, together with desired patterns characterizing the behaviour of strong groups. The goal is to enable the groups and their facilitators to see relevant aspects of the group's operation and provide feedback if these are more likely to be associated with positive or negative outcomes and where the problems are. We explore how useful mirror information can be extracted via a theory-driven approach and a range of clustering and sequential pattern mining. The context is a senior software development project where students use the collaboration tool TRAC. We extract patterns distinguishing the better from the weaker groups and get insights in the success factors. The results point to the importance of leadership and group interaction, and give promising indications if they are occurring. Patterns indicating good individual practices were also identified. We found that some key measures can be mined from early data. The results are promising for advising groups at the start and early identification of effective and poor practices, in time for remediation.

. A density-based unsupervised clustering approach for detecting natural patterns in data further denoted as NP is presented, and its performance is illustrated for data sets with different types of clusters. NP works for arbitrary clusters, is a single-scan technique, requires no presumptions regarding data distribution and requires only one input parameter, which describes the minimal number of objects, considered as cluster. Moreover, a comparison of NP with partitioning approaches is demonstrated. NP can be applied not only for data clustering, but also for the identification of outliers. q 2001 Elsevier Science B.V. All rights reserved.

We present a novel optimization framework for unsupervised texture segmentation that relies on statistical tests as a measure of homogeneity. Texture segmentation is formulated as a data clustering problem based on sparse proximity data. Dissimilarities of pairs of textured regions are computed from a multi scale Gabor lter image representation. We discuss and compare a class of clustering objective functions which is systematically derived from invariance principles. As a general optimization framework we propose deterministic annealing based on a mean eld approximation. The canonical way to derive clustering algorithms within this framework as well as an e cient implementation of mean eld annealing and the closely related Gibbs sampler are presented. We apply both annealing variants to Brodatz like micro texture mixtures and real word images.

Automatic human motion tracking in video sequences is one of the most frequently tackled tasks in computer vision community. The goal of human motion capture is to estimate the joints angles of human body at any time. However, this is one of the most challenging problem in computer vision and pattern recognition due to the high-dimensional search space, self-occlusion, and high variability in human appearance. Several approaches have been proposed in the literature using different techniques. However, conventional approaches such as stochastic particle filtering have shortcomings in computational cost, slowness of convergence, suffers from the curse of dimensionality and demand a high number of evaluations to achieve accurate results. Particle swarm optimization (PSO) is a population-based globalized search algorithm which has been successfully applied to address human motion tracking problem and produced better results in high-dimensional search space. This paper presents a systematic literature survey on the PSO algorithm and its variants to human motion tracking. An attempt is made to provide a guide for the researchers working in the field of PSO based human motion tracking from video sequences. Additionally, the paper also presents the performance of various model evaluation search strategies within PSO tracking framework for 3D pose tracking.

We study the spatial structure and sub-structure of regions rich in Hipparcos stars with blue BT − VT colours. These regions, which comprise large stellar complexes, OB associations, and young open clusters, are tracers of on-going star formation in the Galaxy. The DBSCAN (Density-Based Spatial Clustering of Applications with Noise) data clustering algorithm is used to look for spatial overdensities of early-type stars. Once an overdensity, "agglomerate", is identified, we carry out a data and bibliographic compilation of their star member candidates. The actual membership in agglomerate of each early-type star is studied based on its heliocentric distance, proper motion, and previous spectro-photometric information. We identify 35 agglomerates of early-type Hipparcos stars. Most of them are associated to previously known clusters and OB associations. The previously unknown P Puppis agglomerate is subject of a dedicated study with Virtual Observatory tools. It is actually a new, nearby, young open cluster (d ∼ 470 pc, age ∼ 20 Ma) with a clear radial density gradient. We list P Puppis and other six agglomerates (including NGC 2451 A, vdBH 23, and Trumpler 10) as new sites for substellar searches because of their youth, closeness, and spatial density. We investigate in detail the sub-structure in the Orion, CMa-Pup and Pup-Vel OB complexes ("super-agglomerates"). We confirm or discover some stellar overdensities in the Orion complex, like the 25 Ori group, the Horsehead region (including the σ Orionis cluster), and the η Orionis agglomerate. Finally, we derive accurate parallactic distances to the Pleiades, NGC 2451 A, and IC 2391, describe several field early-type stars at d < 200 pc, and discuss the incompleteness of our search.

In this paper, we propose a new approach of fault detection and diagnosis combining a Neural Nonlinear Principal Component Analysis (NNLPCA) and Partial Least Square (PLS). We have made a comparative study between the Linear Principal Component Analysis (LPCA) and Nonlinear Principal Component Analysis (NLPCA) to monitor a manufacturing process. This study has shown the capability of NLPCA in explaining nonlinear correlations in the process data. The traditional LPCA is limited to complex nonlinear systems; therefore, an adaptive NLPCA based on an improved training auto-associative neural network is presented. The proposed approach is applied to fault detection of a manufacturing process. The performance of the proposed approach is then illustrated and compared to those of classic LPCA.

A resampling scheme for clustering with similarity to bootstrap aggregation (bagging) is presented. Bagging is used to improve the quality of pathbased clustering, a data clustering method that can extract elongated structures from data in a noise robust way. The results of an agglomerative optimization method are influenced by small fluctuations of the input data. To increase the reliability of clustering solutions, a stochastic resampling method is developed to infer consensus clusters. A related reliability measure allows us to estimate the number of clusters, based on the stability of an optimized cluster solution under resampling. The quality of path-based clustering with resampling is evaluated on a large image dataset of human segmentations.

Clustering techniques have received attention in many fields of study such as engineering, medicine, biology and data mining. The aim of clustering is to collect data points. The K-means algorithm is one of the most common techniques used for clustering. However, the results of K-means depend on the initial state and converge to local optima. In order to overcome local optima obstacles, a lot of studies have been done in clustering. This paper presents an efficient hybrid evolutionary optimization algorithm based on combining Modify Imperialist Competitive Algorithm (MICA) and K-means (K), which is called K-ICA, for optimum clustering N objects into K clusters. The new Hybrid K-ICA algorithm is tested on several data sets and its performance is compared with those of MICA, ACO, PSO, Simulated Annealing (SA), Genetic Algorithm (GA), Tabu Search (TS), Honey Bee Mating Optimization (HBMO) and K-means. The simulation results show that the proposed evolutionary optimization algorithm is robust and suitable for handling data clustering.

A modified cluster analysis method has been developed to identify spatial patterns of planetary flow regimes and to study transitions between them. This method has been applied first to a simple deterministic model and second to northern hemisphere (NH) 500-mbar data. The dynamical model is governed by the fully nonlinear, equivalent-barotropic vorticity equation on the sphere. Clusters of points in the model's phase space are associated either with a few persistent events or with many transient events. Two stationary clusters have patterns similar to unstable stationary model solutions, zonal or blocked. Transient clusters of wave trains serve as way stations between the stationary ones. For the NH data, cluster analysis was carried out in the subspace of the first seven empirical orthogonal functions (EOFs). Stationary clusters are found in the low-frequency band of more than 10 days, and transient clusters are found in the band-pass frequency window between 2.5 and 6 days. In the low-frequency band, three pairs of clusters determine EOFs 1, 2, and 3, respectively. They exhibit well-known regional features, such as blocking, the Pacific/North American (PNA) pattern, and wave trains. Both model and low-pass data show strong bimodality. Clusters in the band-pass window show wavetrain patterns in the two jet exit regions.

Background-Advances in multi-parameter flow cytometry (FCM) now allow for the independent detection of larger numbers of fluorochromes on individual cells, generating data with increasingly higher dimensionality. The increased complexity of these data has made it difficult to identify cell populations from high-dimensional FCM data using traditional manual gating strategies based on single-color or two-color displays. Methods-To address this challenge, we developed a novel program, FLOCK (FLOw Clustering without K), that uses a density-based clustering approach to algorithmically identify biologically relevant cell populations from multiple samples in an unbiased fashion, thereby eliminating operator-dependent variability. Results-FLOCK was used to objectively identify seventeen distinct B cell subsets in a human peripheral blood sample and to identify and quantify novel plasmablast subsets responding transiently to tetanus and other vaccinations in peripheral blood. FLOCK has been implemented in the publically available Immunology Database and Analysis Portal-ImmPort (http://www.immport.org) for open use by the immunology research community. Conclusions-FLOCK is able to identify cell subsets in experiments that use multi-parameter flow cytometry through an objective, automated computational approach. The use of algorithms like FLOCK for FCM data analysis obviates the need for subjective and labor intensive manual gating to identify and quantify cell subsets. Novel populations identified by these computational approaches can serve as hypotheses for further experimental study.

A fuzzy rule can have the shape of an ellipsoid in the input-output state space of a system. Then an additive fuzzy system approximates a function by covering its graph with ellipsoidal rule patches. It averages rule patches that overlap. The best fuzzy rules cover the extrema or bumps in the function. Neural or statistical clustering systems can approximate the unknown fuzzy rules from training data. Neural systems can then both tune these rules and add rules to improve the function approximation. We use a hybrid neural system that combines unsupervised and supervised learning to find and tune the rules in the form of ellipsoids. Unsupervised competitive learning finds the first-order and second-order statistics of clusters in the training data. The covariance matrix of each cluster gives an ellipsoid centered at the vector or centroid of the data cluster. The supervised neural system learns with gradient descent. It locally minimizes the mean-squared error of the fuzzy function approximation. In the hybrid system unsupervised learning initializes the gradient descent. The hybrid system tends to give a more accurate function approximation than does the lone unsupervised or supervised system. We found a closed-form model for the optimal rules when only the centroids of the ellipsoids change. We used numerical techniques to find the optimal rules in the general case.

Optimization based pattern discovery has emerged as an important field in knowledge discovery and data mining (KDD), and has been used to enhance the efficiency and accuracy of clustering, classification, association rules and outlier detection. Cluster analysis, which identifies groups of similar data items in large datasets, is one of its recent beneficiaries. The increasing complexity and large amounts of data in the datasets have seen data clustering emerge as a popular focus for the application of optimization based techniques. Different optimization techniques have been applied to investigate the optimal solution for clustering problems. Swarm intelligence (SI) is one such optimization technique whose algorithms have successfully been demonstrated as solutions for different data clustering domains. In this paper we investigate the growth of literature in SI and its algorithms, particularly Particle Swarm Optimization (PSO). This paper makes two major contributions. Firstly, it provides a thorough literature overview focusing on some of the most cited techniques that have been used for PSO-based data clustering. Secondly, we analyze the reported results and highlight the performance of different techniques against contemporary clustering techniques. We also provide an brief overview of our PSO-based hierarchical clustering approach (HPSO-clustering) and compare the results with traditional hierarchical agglomerative clustering (HAC), K-means, and PSO clustering.

We describe the use of principle component analysis (PCA) to serve as a prefilter for cluster analysis or pattern récognition analysis of soft x-ray spectromicroscopy data. Cluster analysis provides a method to group régions with common spectral features even if no prior knowledge of their spectra is available, such as in biology or environmental science.

Conventional clustering algorithms utilize a single criterion that may not conform to the diverse shapes of the underlying clusters. We offer a new clustering approach that uses multiple clustering objective functions simultaneously. The proposed multiobjective clustering is a two-step process. It includes detection of clusters by a set of candidate objective functions as well as their integration into the target partition. A key ingredient of the approach is a cluster goodness function that evaluates the utility of multiple clusters using re-sampling techniques. Multiobjective data clustering is obtained as a solution to a discrete optimization problem in the space of clusters. At meta-level, our algorithm incorporates conflict resolution techniques along with the natural data constraints. An empirical study on a number of artificial and real-world data sets demonstrates that multiobjective data clustering leads to valid and robust data partitions.

We present efficient fixed-parameter algorithms for the NP-complete edge modification problems Cluster Editing and Cluster Deletion. Here, the goal is to make the fewest changes to the edge set of an input graph such that the new graph is a vertex-disjoint union of cliques. Allowing up to k edge additions and deletions (Cluster Editing), we solve this problem in O(2.27 k + |V | 3 ) time; allowing only up to k edge deletions (Cluster Deletion), we solve this problem in O(1.77 k + |V | 3 ) time. The key ingredients of our algorithms are two easy to implement bounded search tree algorithms and an efficient polynomial-time reduction to a problem kernel of size O(k 3 ). This improves and complements previous work. Finally, we discuss further improvements on search tree sizes using computer-generated case distinctions.

We consider the problem of determining the structure of high-dimensional data without prior knowledge of the number of clusters. Data are represented by a finite mixture model based on the generalized Dirichlet distribution. The generalized Dirichlet distribution has a more general covariance structure than the Dirichlet distribution and offers high flexibility and ease of use for the approximation of both symmetric and asymmetric distributions. This makes the generalized Dirichlet distribution more practical and useful. An important problem in mixture modeling is the determination of the number of clusters. Indeed, a mixture with too many or too few components may not be appropriate to approximate the true model. Here, we consider the application of the minimum message length (MML) principle to determine the number of clusters. The MML is derived so as to choose the number of clusters in the mixture model that best describes the data. A comparison with other selection criteria is performed. The validation involves synthetic data, real data clustering, and two interesting real applications: classification of Web pages, and texture database summarization for efficient retrieval.

Home literacy surveys were collected from the primary caregiver of 1,044 2-to 5-year-old children (M = 49.32 months, SD = 9.36) representing a wide range of socioeconomic backgrounds and types of early educational programs or child care. The caregivers completed survey questions on family background and home literacy activities and materials. Hierarchical cluster analyses performed on three randomly constituted subsamples and on the total sample revealed that a threecluster solution best fit the data. Clusters differed on frequency of shared reading and literacy teaching activities with clusters representing caregivers either low or high on all behaviors or low on shared reading behaviors but high on literacy teaching behaviors. Cluster membership was significantly related to socioeconomic status, family living circumstances, caregiver stress, and caregiver reading ability. Results support a model of home literacy behavior informed not only by knowledge and resources but also by parental beliefs and deliberate choices. The cumulative research of the past several decades has led to an increasingly sophisticated understanding of the early childhood underpinnings of reading ability (e.g.

Microarray analysis using clustering algorithms can suffer from lack of inter-method consistency in assigning related gene-expression profiles to clusters. Obtaining a consensus set of clusters from a number of clustering methods should improve confidence in gene-expression analysis. Here we introduce consensus clustering, which provides such an advantage. When coupled with a statistically based gene functional analysis, our method allowed the identification of novel genes regulated by NFκB and the unfolded protein response in certain B-cell lymphomas.

An electronic nose (e-nose), the Cyrano Sciences&#39; Cyranose 320, comprising an array of thirty-two polymer carbon black composite sensors has been used to identify six species of bacteria responsible for eye infections when present at a range of concentrations in saline solutions. Readings were taken from the headspace of the samples by manually introducing the portable e-nose system into a sterile glass containing a fixed volume of bacteria in suspension. Gathered data were a very complex mixture of different chemical compounds. Linear Principal Component Analysis (PCA) method was able to classify four classes of bacteria out of six classes though in reality other two classes were not better evident from PCA analysis and we got 74% classification accuracy from PCA. An innovative data clustering approach was investigated for these bacteria data by combining the 3-dimensional scatter plot, Fuzzy C Means (FCM) and Self Organizing Map (SOM) network. Using these three data clustering...

Dimension reduction in today's vector space based information retrieval system is essential for improving computational efficiency in handling massive amounts of data. A mathematical framework for lower dimensional representation of text data in vector space based information retrieval is proposed using minimization and a matrix rank reduction formula. We illustrate how the commonly used Latent Semantic Indexing based on the Singular Value Decomposition (LSI/SVD) can be derived as a method for dimension reduction from our mathematical framework. Then two new methods for dimension reduction based on the centroids of data clusters are proposed and shown to be more efficient and effective than LSI/SVD when we have a priori information on the cluster structure of the data. Several advantages of the new methods in terms of computational efficiency and data representation in the reduced space, as well as their mathematical properties are discussed. Experimental results are presented to illustrate the effectiveness of our methods on certain classification problems in a reduced dimensional space. The results indicate that for a successful lower dimensional representation of the data, it is important to incorporate a priori knowledge in the dimension reduction algorithms.

Proterozoic subduction metasomatism subcontinental lithospheric mantle post-collisional ore deposit lead isotopes laser-ablation ICP-MS The global supply of Mo and much of Cu and Au comes from porphyry-type ore deposits associated with hydrous magmas of broadly calc-alkaline composition, thought to be generated by contemporaneous subduction zone processes. Molybdenum is generally considered to be derived from the continental crust while Cu and Au are sourced in the mantle wedge above subducting slabs. Here we show that neither contemporaneous subduction nor derivation of Mo from crustal sources is required to explain the genesis of porphyry-Cu-Mo-Au deposits on Proterozoic lithosphere in the eastern Rocky Mountains. Uniform Pb isotope ratios measured by LA-MC-ICP-MS in individual fluid inclusions from distinct Cu-Au and later Mo ore-forming stages at Bingham Canyon, USA, demonstrate a common metal source. Uranogenic Pb isotope ratios are particularly non-radiogenic (17.494 b 206 Pb/ 204 Pb b 17.534; 15.553 b 207 Pb/ 204 Pb b 15.588) and plot to the left of the geochron and above the mantle Pb evolution line. In 207 Pb/ 206 Pb vs. 208 Pb/ 206 Pb space, the fluid Pb isotope data cluster at the non-radiogenic end of a mixing line described by N 80 feldspar data from igneous rocks intimately associated with magmatic-hydrothermal ore formation, which extends to modern depleted mantle or upper crust. Forward Monte Carlo simulations require three events for the U-Th-Pb isotope evolution of the fluid: (1) Late Archean formation of enriched crust is followed by (2) preferential extraction of Pb from this aged crust into a subduction fluid characterized by drastically reduced U/Pb that metasomatized lithospheric mantle at ∼ 1.8 Ga. This mantle reservoir then evolved to produce the retarded uranogenic Pb isotope signatures of the Bingham Canyon Cu-Mo-Au deposit in the Cenozoic (3). Similarly retarded uranogenic Pb isotope data characterize the giant porphyry-Mo and Climax-type Mo deposits of Henderson, Questa, Butte, and SE Arizona that occur in Proterozoic sutures of the central and eastern Rocky Mountains. We propose that Cenozoic melting of subcontinental lithospheric mantle metasomatized by subduction fluids during early Proterozoic amalgamation of terranes to the Wyoming Craton provides the metal endowment and subduction flavour to the giant magmatic-hydrothermal Cu-Mo-Au ore deposits in western North America, which together constitute the world's major molybdenum ore province.