There is a family of well-known external clustering validity indexes to measure the degree of compatibility or similarity between two hard partitions of a given data set, including partitions with different numbers of categories. A unified, fully equivalent set-theoretic formulation for an important class of such indexes was derived and extended to the fuzzy domain in a previous work by the author [Campello, R.J.G.B., 2007. A fuzzy extension of the Rand index and other related indexes for clustering and classification assessment. Pattern Recognition Lett., 28, 833-841]. However, the proposed fuzzy set-theoretic formulation is not valid as a general approach for comparing two fuzzy partitions of data. Instead, it is an approach for comparing a fuzzy partition against a hard referential partition of the data into mutually disjoint categories. In this paper, generalized external indexes for comparing two data partitions with overlapping categories are introduced. These indexes can be used as general measures for comparing two partitions of the same data set into overlapping categories. An important issue that is seldom touched in the literature is also addressed in the paper, namely, how to compare two partitions of different subsamples of data. A number of pedagogical examples and three simulation experiments are presented and analyzed in details. A review of recent related work compiled from the literature is also provided. (c) 2010 Elsevier B.V. All rights reserved.; Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq); Brazilian National Research Council (CNPq); Research Foundation of the State of Sao Paulo (Fapesp)[06/50231-5]; Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP); Research Foundation of the State of Sao Paulo (Fapesp)[301063/2007-9]; Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Introdução: A disfunção erétil (DE) afeta cerca da metade da população masculina adulta com idade superior a 40 anos no Brasil e no mundo. Relacionam-se ao envelhecimento e a doenças como diabetes, problemas cardiovasculares e prostáticos. Para quantificar e avaliar a resposta a tratamento para DE existem questionários específicos, a maioria desenvolvido e validado na língua inglesa. Entre esses instrumentos destacam-se o Índice Internacional de Função Erétil (IIEF) e o Quality of Erection Questionnaire (QEQ). Além dos instrumentos específicos para avaliar a sexualidade masculina, é importante considerar o impacto da DE e seus tratamentos na qualidade de vida geral desses indivíduos. Questionário de Qualidade de Vida Rand 36 (RAND) é um instrumento sensível e reprodutível na detecção de alterações da QV. Além da tradução, são necessárias adaptação cultural e validação de instrumentos desenvolvidos em inglês, para adequá-los à cultura brasileira e garantir propriedades psicométricas similares a versão no idioma original. Objetivos: Traduzir, adaptar culturalmente e validar instrumento QEQ para homens brasileiros e comparar seus escores com os do IIEF e os do RAND 36-Item Health Survey. Pacientes e Métodos: A versão em inglês do QEQ foi submetida a tradução...

Mestrado em Prospecção e Análise de Dados; A zona turística rural no norte de Portugal é uma das áreas escolhidas por alguns turistas que visitam o nosso País. A auto-imagem desses turistas pode aferir-se mediante atributos capazes de caracterizar também o destino visitado. Neste trabalho aplica-se a metodologia de estimação de modelos de mistura finita para constituição e caracterização dos segmentos de turistas com base em atributos de autoimagem. Na modelação atende-se ao tipo de escala – Semântica Diferencial – de dois modos distintos: 1 – as variáveis base são modeladas como métricas e usada uma mistura de distribuições normais; 2 – as variáveis base são modeladas como ordinais e usada uma mistura de multinomiais. Aplica-se a metodologia de validação cruzada com o auxílio dos índices de concordância para comparar a estabilidade das soluções alternativas de agrupamento, com base na modelação das variáveis como métricas ou ordinais. No trabalho desenvolvido o agrupamento com base nas variáveis ordinais tem uma maior estabilidade. Este agrupamento é seleccionado e caracterizado.; The touristic rural zone in the north of Portugal is one of the areas chosen by some tourists who visit our Country. The auto-image of these tourists can be checked by means of attributes able to also characterize the visited destination. In this paper...

Purpose To validate the Quality of Erection Questionnaire (QEQ) considering Brazilian social-cultural aspects. Materials and Methods To determine equivalence between the Portuguese and the English QEQ versions, the Portuguese version was back-translated by two professors who are native English speakers. After language equivalence had been determined, urologists considered the QEQ Portuguese version suitable. Men with self-reported erectile dysfunction (ED) and infertile men who had a stable sexual relationship for at least 6 months were invited to answer the QEQ, the International Index of Erectile Function (IIEF) and the RAND 36-Item Health Survey (RAND-36). The questionnaires were presented together and answered without help in a private room. Internal consistency (Cronbach’s α), test-retest reliability (Spearman), convergent validity (Spearman correlation) coefficients and known-groups validity (the ability of the QEQ Portuguese version to differentiate erectile dysfunction severity groups) were assessed. Results We recruited 197 men (167 ED patients and 30 non-ED patients), mean age of 53.3 and median of 55.5 years (23-82 years). The Portuguese version of the QEQ had high internal consistency (Cronbach α=0.93)...

A moment closure model of sexually transmitted disease spread through a concurrent partnership network is developed. The model employs pair approximations of higher-order correlations to derive equations of motion in terms of numbers of pairs and singletons. The model is derived from an underlying stochastic process of partnership network formation and disease transmission. The model is analysed numerically; and the final size and time evolution are considered for various levels of concurrency, as measured by the concurrency index kappa3 of Kretzschmar and Morris. Additionally, a new way of calculating R0 for spatial network models is developed. It is found that concurrency significantly increases R0 and the final size of a sexually transmitted disease, with some interesting exceptions.

The relationship between primary open-angle glaucoma (POAG) and cerebrospinal fluid pressure (CSFP) is explored. Increasing body mass index (BMI), which is protective against POAG, is associated with an increase in CSFP. This association may explain how an elevated BMI decreases the risk for POAG.

Complex diseases are typically caused by combinations of molecular disturbances that vary widely among different patients. Endophenotypes, a combination of genetic factors associated with a disease, offer a simplified approach to dissect complex trait by reducing genetic heterogeneity. Because molecular dissimilarities often exist between patients with indistinguishable disease symptoms, these unique molecular features may reflect pathogenic heterogeneity. To detect molecular dissimilarities among patients and reduce the complexity of high-dimension data, we have explored an endophenotype-identification analytical procedure that combines non-negative matrix factorization (NMF) and adjusted rand index (ARI), a measure of the similarity of two clusterings of a data set. To evaluate this procedure, we compared it with a commonly used method, principal component analysis with k-means clustering (PCA-K). A simulation study with gene expression dataset and genotype information was conducted to examine the performance of our procedure and PCA-K. The results showed that NMF mostly outperformed PCA-K. Additionally, we applied our endophenotype-identification analytical procedure to a publicly available dataset containing data derived from patients with late-onset Alzheimer’s disease (LOAD). NMF distilled information associated with 1...

Background: To assess whether the individual housing-based socioeconomic status (SES) measure termed HOUSES was associated with post-myocardial infarction (MI) mortality. Methods: The study was designed as a population-based cohort study, which compared post-MI mortality among Olmsted County, Minnesota, USA, residents with different SES as measured by HOUSES using Cox proportional hazards models. Subjects’ addresses at index date of MI were geocoded to real property data to formulate HOUSES (a z-score for housing value, square footage, and numbers of bedrooms and bathrooms). Educational levels were used as a comparison for the HOUSES index. Results: 637 of the 696 eligible patients with MI (92%) were successfully geocoded to real property data. Post-MI survival rates were 60% (50–72), 78% (71–85), 72% (60–87), and 87% (81–93) at 2 years for patients in the first (the lowest SES), second, third, and fourth quartiles of HOUSES, respectively (p < 0.001). HOUSES was associated with post-MI all-cause mortality, controlling for all variables except age and comorbidity (p = 0.036) but was not significant after adjusting for age and comorbidity (p = 0.24). Conclusions: Although HOUSES is associated with post-MI mortality, the differential mortality rates by HOUSES were primarily accounted for by age and comorbid conditions. HOUSES may be useful for health disparities research concerning cardiovascular outcomes...

Vibrio parahaemolyticus is the leading cause of seafood-borne gastroenteritis outbreaks. To track the source of these diseases in a timely manner, a high throughput typing method is critical. We hereby describe a novel genotyping method for V. parahaemolyticus, termed multilocus melt typing (MLMT), based on multilocus sequence typing (MLST). MLMT utilizes melting curve analysis to interrogate the allelic types of a set of informative single nucleotide polymorphisms (SNPs) derived from the housekeeping genes used in MLST. For each SNP, one allelic type generates distinct Tm values, which are converted into a binary code. Multiple SNPs thus generate a series of binary codes, forming a melt type (MT) corresponding with a sequence type (ST) of MLST. Using a set of 12 SNPs, the MLMT scheme could resolve 218 V.parahaemolyticus isolates into 50 MTs corresponding with 56 STs. The discriminatory power of MLMT and MLST was similar with Simpson’s index of diversity of 0.638 and 0.646, respectively. The global (adjusted Rand index = 0.982) and directional congruence (adjusted Wallace coefficient, MT→ST = 0.965; ST→MT = 1.000) between the two typing approaches was high. The entire procedure of MLMT could be finished within 3 h with negligible hands on time in a real-time PCR machine. We conclude that MLMT provides a reliable and efficient approach for V. parahaemolyticus genotyping and might also find use in other pathogens.

A técnica de análise de agrupamento (clustering analysis) é uma ferramenta importante na pesquisa científica, podendo ser utilizada em diversas áreas do conhecimento tais como medicina, biologia e estatística. Agrupar dados é uma forma de refletir a estrutura interna dos dados e identificar classes presentes nesses agrupamentos, de modo que haja homogeneidade dentro das mesmas classes e heterogeneidade entre classes diferentes. Existem vários métodos de agrupamentos utilizados para encontrar o particionamento ótimo, dentre estes pode-se destacar: os métodos hierárquicos, métodos baseados em teorias dos grafos e métodos baseados em função objetivo. Neste trabalho foi utilizado o algoritmo baseado na função objetivo Fuzzy C-Means em conjunto com a técnica de reamostragem bootstrap. A ideia é variar o índice de nebulosidade para encontrar a melhor faixa de valores a ser utilizada para a classificação dos dados e consequentemente obtenção de melhores particionamentos. A qualidade da classificação é baseada em medidas de comparação tradicionais tais como Classificação Cruzada (Acc), F1, Hubert (Hub), Jaccard, Índice Randômico (Rand) e Fowlkes and Mallows (Fowlkes). As bases de dados utilizadas foram a Iris...

The correct quantification of mixing is of utmost importance for modeling reactive transport in porous media and for assessing the fate and transport of contaminants in the subsurface. In this work we mainly focus on two-dimensional systems under steady-state flow conditions with a continuous, constant linear source. We investigate both conservative and reactive transport in homogeneous and heterogeneous porous media and hereby, consider laboratory bench-scale experiments as well as numerical models at the Darcy and at the field scales. Under the specified conditions the most important mixing processes occur in the transverse direction perpendicular to the main flow. The leading question behind this work is hence, how we can properly quantify mixing in the transverse direction at the Darcy and at the field scales. Quasi two-dimensional laboratory bench-scale experiments have been performed in order to study: i) the effects of flow focusing in high-permeability zones on transverse mixing for conservative and reactive solute transport; ii) the compound-specific behavior of mechanical dispersion at the Darcy scale; iii) the effects of compound-specific local transverse dispersion coefficients on isotope fractionation for conservative and reactive (i.e. aerobic and anaerobic biodegradation) transport. Numerical models have been used to quantitatively interpret the outcomes of the laboratory experiments and to study different measures of dilution and mixing for conservative and reactive solute transport at the Darcy and at the field scales. Different scenarios have been investigated at the field scale...

We introduce SAGRA (Split And Group Recombining Algorithm), a cluster analysis methodology which split the data set into small homogeneous groups and later recombine those groups using Bayes factors. We compare the performance of SAGRA with other three cluster analysis algorithms: SAR, M-clust and K-means, using five quality measures: Purity, number of groups, Rand index, adjusted Rand index, and F1, over four different data configurations. Results indicate that the SAGRA algorithm obtain consistently similar or better indexes than the other algorithms over all measures and data configurations; Work partially supported by Spanish Ministry of Science and Innovation, research projects SEJ2007-64500 and ECO2012-38442

In genetic association studies, unaccounted population stratification can cause spurious associations in a discovery process of identifying disease-associated genetic markers. In such a situation, prior information is often available for some subjects' population identities. To leverage the additional information, we propose a semi-supervised clustering approach for detecting population stratification. This approach maintains the advantages of spectral clustering, while is integrated with the additional identity information, leading to sharper clustering performance. To demonstrate utility of our approach, we analyze a whole-genome sequencing dataset from the 1000 Genomes Project, consisting of the genotypes of 607 individuals sampled from three continental groups involving 10 subpopulations. This is compared against a semi-supervised spectral clustering method, in addition to a spectral clustering method, with the known subpopulation information by the Rand index and an adjusted Rand (ARand) index. The numerical results suggest that the proposed method outperforms its competitors in detecting population stratification.

The popularity of Internet usage although increases exponentially, it is incapable of providing the security for exchange of confidential data between the users. As a result, several cryptosystems for encryption of data and images have been developed for secured transmission over Internet. In this work, a scheme for Image encryption/decryption based on Vector Quantization (VQ) has been proposed that concurrently encodes the images for compression and shuffles the codebook and the index matrix using pseudorandom sequences for encryption. The processing time of the proposed scheme is much less than the other cryptosystems, because it does not use any traditional cryptographic operations, and instead it performs swapping between the contents of the codebook with respect to a random sequence, which resulted an indirect shuffling of the contents of the index matrix. It may be noted that the security of the proposed cryptosystem depends on the generation and the exchange of the random sequences used. Since the generation of truly random sequences are not practically feasible, we simulate the proposed scheme using MATLAB, where its operators like rand(method, seed), randperm(n) has been used to generate pseudorandom sequences and it has been seen that the proposed cryptosystem shows the expected performance.

Clustering techniques are widely used in Web Usage Mining to capture similar interests and trends among users accessing a Web site. For this purpose, web access logs generated at a particular web site are preprocessed to discover the user navigational sessions. Clustering techniques are then applied to group the user session data into user session clusters, where intercluster similarities are minimized while the intra cluster similarities are maximized. Since the application of different clustering algorithms generally results in different sets of cluster formation, it is important to evaluate the performance of these methods in terms of accuracy and validity of the clusters, and also the time required to generate them, using appropriate performance measures. This paper describes various validity and accuracy measures including Dunn's Index, Davies Bouldin Index, C Index, Rand Index, Jaccard Index, Silhouette Index, Fowlkes Mallows and Sum of the Squared Error (SSE). We conducted the performance evaluation of the following clustering techniques: k-Means, k-Medoids, Leader, Single Link Agglomerative Hierarchical and DBSCAN. These techniques are implemented and tested against the Web user navigational data. Finally their performance results are presented and compared.

In comparing clustering partitions, Rand index (RI) and Adjusted Rand index (ARI) are commonly used for measuring the agreement between the partitions. Both these external validation indexes aim to analyze how close is a cluster to a reference (or to prior knowledge about the data) by counting corrected classified pairs of elements. When the aim is to evaluate the solution of a fuzzy clustering algorithm, the computation of these measures require converting the soft partitions into hard ones. It is known that different fuzzy partitions describing very different structures in the data can lead to the same crisp partition and consequently to the same values of these measures. \noindent We compare the existing approaches to evaluate the external validation criteria in fuzzy clustering and we propose an extension of the ARI for fuzzy partitions based on the normalized degree of concordance. Through use of simulated data, we analyze and evaluate the performance of our proposal.; Comment: Work presented at the International Federation of Classification Society (IFCS), Bologna (Italy), 2015

Images can be segmented by first using a classifier to predict an affinity graph that reflects the degree to which image pixels must be grouped together and then partitioning the graph to yield a segmentation. Machine learning has been applied to the affinity classifier to produce affinity graphs that are good in the sense of minimizing edge misclassification rates. However, this error measure is only indirectly related to the quality of segmentations produced by ultimately partitioning the affinity graph. We present the first machine learning algorithm for training a classifier to produce affinity graphs that are good in the sense of producing segmentations that directly minimize the Rand index, a well known segmentation performance measure. The Rand index measures segmentation performance by quantifying the classification of the connectivity of image pixel pairs after segmentation. By using the simple graph partitioning algorithm of finding the connected components of the thresholded affinity graph, we are able to train an affinity classifier to directly minimize the Rand index of segmentations resulting from the graph partitioning. Our learning algorithm corresponds to the learning of maximin affinities between image pixel pairs...

In this thesis, we propose a fast unsupervised multiresolution color image segmentation algorithm which takes advantage of gradient information in an adaptive and progressive framework. This gradient-based segmentation method is initialized by a vector gradient calculation on the full resolution input image in the CIE L*a*b* color space. The resultant edge map is used to adaptively generate thresholds for classifying regions of varying gradient densities at different levels of the input image pyramid, obtained through a dyadic wavelet decomposition scheme. At each level, the classification obtained by a progressively thresholded growth procedure is combined with an entropy-based texture model in a statistical merging procedure to obtain an interim segmentation. Utilizing an association of a gradient quantized confidence map and non-linear spatial filtering techniques, regions of high confidence are passed from one level to another until the full resolution segmentation is achieved. Evaluation of our results on several hundred images using the Normalized Probabilistic Rand (NPR) Index shows that our algorithm outperforms state-of the art segmentation techniques and is much more computationally efficient than its single scale counterpart...

This paper seeks to examine the dynamic causal relations between the two major financial assets, stock prices of the US and South Africa and the rand/US$ exchange rate. The study uses a mixed bag of time series approaches such as cointegration, Granger causality, impulse response functions and forecasting error variance decompositions. The paper identifies a bi-directional causality from the Standard & Poor's 500 stock price index to the rand/US$ exchange rate in the Granger sense. It was also found that the Standard & Poor's stock price index accounts for a significant portion of the variations in the Johannesburg Stock Exchange's All Share index. Thus, while causality in the Granger sense could not be established for the relationship between the price indices of the two stock exchanges it can argued that there is some relationship between them. The results of the study have implications for both business and Government.