Quality indices for (practical) clustering evaluation

Clustering quality or validation indices allow the evaluation of the quality of clustering in order to support the selection of a specific partition or clustering structure in its natural unsupervised environment, where the real solution is unknown or not available. In this paper, we investigate the use of quality indices mostly based on the concepts of clusters` compactness and separation, for the evaluation of clustering results (partitions in particular). This work intends to offer a general perspective regarding the appropriate use of quality indices for the purpose of clustering evaluation. After presenting some commonly used indices, as well as indices recently proposed in the literature, key issues regarding the practical use of quality indices are addressed. A general methodological approach is presented which considers the identification of appropriate indices thresholds. This general approach is compared with the simple use of quality indices for evaluating a clustering solution.

Spatial clustering analysis of the foot-and-mouth disease outbreaks in Mato Grosso do Sul state, Brazil - 2005

In the southern region of Mato Grosso do Sul state, Brazil, a foot-and-mouth disease (FMD) epidemic started in September 2005. A total of 33 outbreaks were detected and 33,741 FMD-susceptible animals were slaughtered and destroyed. There were no reports of FMD cases in other species than bovines. Based on the data of this epidemic, it was carried out an analysis using the K-function and it was observed spatial clustering of outbreaks within a range of 25km. This observation may be related to the dynamics of foot-and-mouth disease spread and to the measures undertaken to control the disease dissemination. The control measures were effective once the disease did not spread to farms more than 47 km apart from the initial outbreaks.

The full Bayesian significance test for mixture models: results in gene expression clustering

Gene clustering is a useful exploratory technique to group together genes with similar expression levels under distinct cell cycle phases or distinct conditions. It helps the biologist to identify potentially meaningful relationships between genes. In this study, we propose a clustering method based on multivariate normal mixture models, where the number of clusters is predicted via sequential hypothesis tests: at each step, the method considers a mixture model of m components (m = 2 in the first step) and tests if in fact it should be m - 1. If the hypothesis is rejected, m is increased and a new test is carried out. The method continues (increasing m) until the hypothesis is accepted. The theoretical core of the method is the full Bayesian significance test, an intuitive Bayesian approach, which needs no model complexity penalization nor positive probabilities for sharp hypotheses. Numerical experiments were based on a cDNA microarray dataset consisting of expression levels of 205 genes belonging to four functional categories, for 10 distinct strains of Saccharomyces cerevisiae. To analyze the method's sensitivity to data dimension, we performed principal components analysis on the original dataset and predicted the number of classes using 2 to 10 principal components. Compared to Mclust (model-based clustering), our method shows more consistent results.

Musical genres: beating to the rhythms of different drums

Online music databases have increased significantly as a consequence of the rapid growth of the Internet and digital audio, requiring the development of faster and more efficient tools for music content analysis. Musical genres are widely used to organize music collections. In this paper, the problem of automatic single and multi-label music genre classification is addressed by exploring rhythm-based features obtained from a respective complex network representation. A Markov model is built in order to analyse the temporal sequence of rhythmic notation events. Feature analysis is performed by using two multi-variate statistical approaches: principal components analysis (unsupervised) and linear discriminant analysis (supervised). Similarly, two classifiers are applied in order to identify the category of rhythms: parametric Bayesian classifier under the Gaussian hypothesis (supervised) and agglomerative hierarchical clustering (unsupervised). Qualitative results obtained by using the kappa coefficient and the obtained clusters corroborated the effectiveness of the proposed method.

Improvements on ICA mixture models for image pre-processing and segmentation

Today several different unsupervised classification algorithms are commonly used to cluster similar patterns in a data set based only on its statistical properties. Specially in image data applications, self-organizing methods for unsupervised classification have been successfully applied for clustering pixels or group of pixels in order to perform segmentation tasks. The first important contribution of this paper refers to the development of a self-organizing method for data classification, named Enhanced Independent Component Analysis Mixture Model (EICAMM), which was built by proposing some modifications in the Independent Component Analysis Mixture Model (ICAMM). Such improvements were proposed by considering some of the model limitations as well as by analyzing how it should be improved in order to become more efficient. Moreover, a pre-processing methodology was also proposed, which is based on combining the Sparse Code Shrinkage (SCS) for image denoising and the Sobel edge detector. In the experiments of this work, the EICAMM and other self-organizing models were applied for segmenting images in their original and pre-processed versions. A comparative analysis showed satisfactory and competitive image segmentation results obtained by the proposals presented herein. (C) 2008 Published by Elsevier B.V.

HYBRID AND INCREMENTAL FUZZY LEARNING FOR HUMAN SKIN DETECTION

In this paper, a framework for detection of human skin in digital images is proposed. This framework is composed of a training phase and a detection phase. A skin class model is learned during the training phase by processing several training images in a hybrid and incremental fuzzy learning scheme. This scheme combines unsupervised-and supervised-learning: unsupervised, by fuzzy clustering, to obtain clusters of color groups from training images; and supervised to select groups that represent skin color. At the end of the training phase, aggregation operators are used to provide combinations of selected groups into a skin model. In the detection phase, the learned skin model is used to detect human skin in an efficient way. Experimental results show robust and accurate human skin detection performed by the proposed framework.

A spectral clustering algorithm for manufacturing cell formation

A graph clustering algorithm constructs groups of closely related parts and machines separately. After they are matched for the least intercell moves, a refining process runs on the initial cell formation to decrease the number of intercell moves. A simple modification of this main approach can deal with some practical constraints, such as the popular constraint of bounding the maximum number of machines in a cell. Our approach makes a big improvement in the computational time. More importantly, improvement is seen in the number of intercell moves when the computational results were compared with best known solutions from the literature. (C) 2009 Elsevier Ltd. All rights reserved.

Gene expression profile of residual breast cancer after doxorubicin and cyclophosphamide neoadjuvant chemotherapy

In breast cancer patients, primary chemotherapy is associated with the same survival benefits as adjuvant chemotherapy. Residual tumors represent a clinical challenge, Lis they may be resistant to additional cycles of the same drugs. Our aim was to identify differential transcripts expressed in residual tumors, after neoadjuvant chemotherapy, that might be related with tumor resistance. Hence, 16 patients with paired tumor samples, collected before and after treatment (4 cycles doxorubicin/cyclophosphamide, AC) had their gene expression evaluated on cDNA microarray slides containing 4,608 genes. Three hundred and eighty-nine genes were differentially expressed (paired Student`s t-test, pFDR<0.01) between pre- and post-chemotherapy samples and among the regulated functions were the JNK cascade and cell death. Unsupervised hierarchical clustering identified one branch comprising exclusively, eight pre-chemotherapy samples and another branch, including the former correspondent eight post-chemotherapy samples and other 16 paired pre/post-chemotherapy samples. No differences in clinical and tumor parameters could explain this clustering. Another group of I I patients with paired samples had expression of selected genes determined by real-time RT-PCR and CTGF and DUSP1 were confirmed more expressed in post- as compared to pre-chemotherapy samples. After neoadjuvant chemotherapy some residual samples may retain their molecular signature while others present significant changes in their gene expression, probably induced by the treatment. CTGF and DUSP1 overexpression in residual samples may be a reflection of resistance to further administration of AC regimen.

Expression of genes related to apoptosis, cell cycle and signaling pathways are independent of TP53 status in urinary bladder cancer cells

Urinary bladder cancer is the fourth most common malignancy in the Western world. Transitional cell carcinoma (TCC) is the most common subtype, accounting for about 90% of all bladder cancers. The TP53 gene plays an essential role in the regulation of the cell cycle and apoptosis and therefore contributes to cellular transformation and malignancy; however, little is known about the differential gene expression patterns in human tumors that present with the wild-type or mutated TP53 gene. Therefore, because gene profiling can provide new insights into the molecular biology of bladder cancer, the present study aimed to compare the molecular profiles of bladder cancer cell lines with different TP53 alleles, including the wild type (RT4) and two mutants (5637, with mutations in codons 280 and 72; and T24, a TP53 allele encoding an in-frame deletion of tyrosine 126). Unsupervised hierarchical clustering and gene networks were constructed based on data generated by cDNA microarrays using mRNA from the three cell lines. Differentially expressed genes related to the cell cycle, cell division, cell death, and cell proliferation were observed in the three cell lines. However, the cDNA microarray data did not cluster cell lines based on their TP53 allele. The gene profiles of the RT4 cells were more similar to those of T24 than to those of the 5637 cells. While the deregulation of both the cell cycle and the apoptotic pathways was particularly related to TCC, these alterations were not associated with the TP53 status.

Characterization of spliced leader genes of Trypanosoma (Megatrypanum) theileri: phylogeographical analysis of Brazilian isolates from cattle supports spatial clustering of genotypes and parity with ribosomal markers

Trypanosoma (Megatrypanum) theileri from cattle and trypanosomes of other artiodactyls form a clade of closely related species in analyses using ribosomal sequences. Analysis of polymorphic sequences of a larger number of trypanosomes from broader geographical origins is required to evaluate the Clustering of isolates as suggested by previous studies. Here, we determined the sequences of the spliced leader (SL) genes of 21 isolates from cattle and 2 from water buffalo from distant regions of Brazil. Analysis of SL gene repeats revealed that the 5S rRNA gene is inserted within the intergenic region. Phylogeographical patterns inferred using SL sequences showed at least 5 major genotypes of T. theileri distributed in 2 strongly divergent lineages. Lineage TthI comprises genotypes IA and IB from buffalo and cattle, respectively, from the Southeast and Central regions, whereas genotype IC is restricted to cattle from the Southern region. Lineage Tth II includes cattle genotypes IIA, which is restricted to the North and Northeast, and IIB, found in the Centre, West, North and Northeast. PCR-RFLP of SL genes revealed valuable markers for genotyping T. theileri. The results of this study emphasize the genetic complexity and corroborate the geographical structuring of T. theileri genotypes found in cattle.

Comparative phylogeography of Trypanosoma cruzi TCIIc: New hosts, association with terrestrial ecotopes, and spatial clustering

We characterized 28 new isolates of Trypanosoma cruzi IIc (TCIIc) of mammals and triatomines from Northern to Southern Brazil, confirming the widespread distribution of this lineage. Phylogenetic analyses using cytochrome b and SSU rDNA sequences clearly separated TCIIc from TCIIa according to terrestrial and arboreal ecotopes of their preferential mammalian hosts and vectors. TCIIc was more closely related to TCIId/e, followed by TCIIa, and separated by large distances from TCIIb and TCI. Despite being indistinguishable by traditional genotyping and generally being assigned to Z3, we provide evidence that TCIIa from South America and TCIIa from North America correspond to independent lineages that circulate in distinct hosts and ecological niches. Armadillos, terrestrial didelphids and rodents, and domestic dogs were found infected by TCIIc in Brazil. We believe that, in Brazil, this is the first description of TCIIc from rodents and domestic dogs. Terrestrial triatomines of genera Panstrongylus and Triatoma were confirmed as vectors of TCIIc. Together, habitat, mammalian host and vector association corroborated the link between TCIIc and terrestrial transmission cycles/ecological niches. Analysis of ITS1 rDNA sequences disclosed clusters of TCIIc isolates in accordance with their geographic origin, independent of their host species. (C) 2009 Elsevier B.V. All rights reserved.

Evolutionary fuzzy clustering of relational data

This paper is concerned with the computational efficiency of fuzzy clustering algorithms when the data set to be clustered is described by a proximity matrix only (relational data) and the number of clusters must be automatically estimated from such data. A fuzzy variant of an evolutionary algorithm for relational clustering is derived and compared against two systematic (pseudo-exhaustive) approaches that can also be used to automatically estimate the number of fuzzy clusters in relational data. An extensive collection of experiments involving 18 artificial and two real data sets is reported and analyzed. (C) 2011 Elsevier B.V. All rights reserved.

Partitions selection strategy for set of clustering solutions

Clustering is a difficult task: there is no single cluster definition and the data can have more than one underlying structure. Pareto-based multi-objective genetic algorithms (e.g., MOCK Multi-Objective Clustering with automatic K-determination and MOCLE-Multi-Objective Clustering Ensemble) were proposed to tackle these problems. However, the output of such algorithms can often contains a high number of partitions, becoming difficult for an expert to manually analyze all of them. In order to deal with this problem, we present two selection strategies, which are based on the corrected Rand, to choose a subset of solutions. To test them, they are applied to the set of solutions produced by MOCK and MOCLE in the context of several datasets. The study was also extended to select a reduced set of partitions from the initial population of MOCLE. These analysis show that both versions of selection strategy proposed are very effective. They can significantly reduce the number of solutions and, at the same time, keep the quality and the diversity of the partitions in the original set of solutions. (C) 2010 Elsevier B.V. All rights reserved.

Investigation of a new GRASP-based clustering algorithm applied to biological data

A large amount of biological data has been produced in the last years. Important knowledge can be extracted from these data by the use of data analysis techniques. Clustering plays an important role in data analysis, by organizing similar objects from a dataset into meaningful groups. Several clustering algorithms have been proposed in the literature. However, each algorithm has its bias, being more adequate for particular datasets. This paper presents a mathematical formulation to support the creation of consistent clusters for biological data. Moreover. it shows a clustering algorithm to solve this formulation that uses GRASP (Greedy Randomized Adaptive Search Procedure). We compared the proposed algorithm with three known other algorithms. The proposed algorithm presented the best clustering results confirmed statistically. (C) 2009 Elsevier Ltd. All rights reserved.

Multi-objective clustering ensemble for gene expression data analysis

In this paper, we present an algorithm for cluster analysis that integrates aspects from cluster ensemble and multi-objective clustering. The algorithm is based on a Pareto-based multi-objective genetic algorithm, with a special crossover operator, which uses clustering validation measures as objective functions. The algorithm proposed can deal with data sets presenting different types of clusters, without the need of expertise in cluster analysis. its result is a concise set of partitions representing alternative trade-offs among the objective functions. We compare the results obtained with our algorithm, in the context of gene expression data sets, to those achieved with multi-objective Clustering with automatic K-determination (MOCK). the algorithm most closely related to ours. (C) 2009 Elsevier B.V. All rights reserved.