Uma Análise Comparativa entre Sistemas de Combinação de Classificadores com Distribuição Vertical dos Dados

In systems that combine the outputs of classification methods (combination systems), such as ensembles and multi-agent systems, one of the main constraints is that the base components (classifiers or agents) should be diverse among themselves. In other words, there is clearly no accuracy gain in a system that is composed of a set of identical base components. One way of increasing diversity is through the use of feature selection or data distribution methods in combination systems. In this work, an investigation of the impact of using data distribution methods among the components of combination systems will be performed. In this investigation, different methods of data distribution will be used and an analysis of the combination systems, using several different configurations, will be performed. As a result of this analysis, it is aimed to detect which combination systems are more suitable to use feature distribution among the components

Uma Análise de métodos de distriubuição de atributos em comitês de classificadores

The objective of the researches in artificial intelligence is to qualify the computer to execute functions that are performed by humans using knowledge and reasoning. This work was developed in the area of machine learning, that it s the study branch of artificial intelligence, being related to the project and development of algorithms and techniques capable to allow the computational learning. The objective of this work is analyzing a feature selection method for ensemble systems. The proposed method is inserted into the filter approach of feature selection method, it s using the variance and Spearman correlation to rank the feature and using the reward and punishment strategies to measure the feature importance for the identification of the classes. For each ensemble, several different configuration were used, which varied from hybrid (homogeneous) to non-hybrid (heterogeneous) structures of ensemble. They were submitted to five combining methods (voting, sum, sum weight, multiLayer Perceptron and naïve Bayes) which were applied in six distinct database (real and artificial). The classifiers applied during the experiments were k- nearest neighbor, multiLayer Perceptron, naïve Bayes and decision tree. Finally, the performance of ensemble was analyzed comparatively, using none feature selection method, using a filter approach (original) feature selection method and the proposed method. To do this comparison, a statistical test was applied, which demonstrate that there was a significant improvement in the precision of the ensembles

Seleção de atributos em comitês de classificadores utilizando algoritmos genéticos

Classifier ensembles are systems composed of a set of individual classifiers and a combination module, which is responsible for providing the final output of the system. In the design of these systems, diversity is considered as one of the main aspects to be taken into account since there is no gain in combining identical classification methods. The ideal situation is a set of individual classifiers with uncorrelated errors. In other words, the individual classifiers should be diverse among themselves. One way of increasing diversity is to provide different datasets (patterns and/or attributes) for the individual classifiers. The diversity is increased because the individual classifiers will perform the same task (classification of the same input patterns) but they will be built using different subsets of patterns and/or attributes. The majority of the papers using feature selection for ensembles address the homogenous structures of ensemble, i.e., ensembles composed only of the same type of classifiers. In this investigation, two approaches of genetic algorithms (single and multi-objective) will be used to guide the distribution of the features among the classifiers in the context of homogenous and heterogeneous ensembles. The experiments will be divided into two phases that use a filter approach of feature selection guided by genetic algorithm

What is the importance of selecting features for non-technical losses identification?

Although non-technical losses automatic identification has been massively studied, the problem of selecting the most representative features in order to boost the identification accuracy has not attracted much attention in this context. In this paper, we focus on this problem applying a novel feature selection algorithm based on Particle Swarm Optimization and Optimum-Path Forest. The results demonstrated that this method can improve the classification accuracy of possible frauds up to 49% in some datasets composed by industrial and commercial profiles. © 2011 IEEE.

New trends in musical genre classification using optimum-path forest

Musical genre classification has been paramount in the last years, mainly in large multimedia datasets, in which new songs and genres can be added at every moment by anyone. In this context, we have seen the growing of musical recommendation systems, which can improve the benefits for several applications, such as social networks and collective musical libraries. In this work, we have introduced a recent machine learning technique named Optimum-Path Forest (OPF) for musical genre classification, which has been demonstrated to be similar to the state-of-the-art pattern recognition techniques, but much faster for some applications. Experiments in two public datasets were conducted against Support Vector Machines and a Bayesian classifier to show the validity of our work. In addition, we have executed an experiment using very recent hybrid feature selection techniques based on OPF to speed up feature extraction process. © 2011 International Society for Music Information Retrieval.

Kappa-PSO-ARTMAP Fuzzy: uma metodologia para detecção de intrusos baseado em seleção de atributos e otimização de parâmetros numa rede neural ARTMAP Fuzzy

Pós-graduação em Engenharia Elétrica - FEIS

Classificação fonética utilizando Boosting e SVM

Para compor um sistema de Reconhecimento Automático de Voz, pode ser utilizada uma tarefa chamada Classificação Fonética, onde a partir de uma amostra de voz decide-se qual fonema foi emitido por um interlocutor. Para facilitar a classificação e realçar as características mais marcantes dos fonemas, normalmente, as amostras de voz são pré- processadas através de um fronl-en'L Um fron:-end, geralmente, extrai um conjunto de parâmetros para cada amostra de voz. Após este processamento, estes parâmetros são insendos em um algoritmo classificador que (já devidamente treinado) procurará decidir qual o fonema emitido. Existe uma tendência de que quanto maior a quantidade de parâmetros utilizados no sistema, melhor será a taxa de acertos na classificação. A contrapartida para esta tendência é o maior custo computacional envolvido. A técnica de Seleção de Parâmetros tem como função mostrar quais os parâmetros mais relevantes (ou mais utilizados) em uma tarefa de classificação, possibilitando, assim, descobrir quais os parâmetros redundantes, que trazem pouca (ou nenhuma) contribuição à tarefa de classificação. A proposta deste trabalho é aplicar o classificador SVM à classificação fonética, utilizando a base de dados TIMIT, e descobrir os parâmetros mais relevantes na classificação, aplicando a técnica Boosting de Seleção de Parâmetros.

Ferramentas e recursos livres para reconhecimento e síntese de voz em português brasileiro

Sistemas de reconhecimento e síntese de voz são constituídos por módulos que dependem da língua e, enquanto existem muitos recursos públicos para alguns idiomas (p.e. Inglês e Japonês), os recursos para Português Brasileiro (PB) ainda são escassos. Outro aspecto é que, para um grande número de tarefas, a taxa de erro dos sistemas de reconhecimento de voz atuais ainda é elevada, quando comparada à obtida por seres humanos. Assim, apesar do sucesso das cadeias escondidas de Markov (HMM), é necessária a pesquisa por novos métodos. Este trabalho tem como motivação esses dois fatos e se divide em duas partes. A primeira descreve o desenvolvimento de recursos e ferramentas livres para reconhecimento e síntese de voz em PB, consistindo de bases de dados de áudio e texto, um dicionário fonético, um conversor grafema-fone, um separador silábico e modelos acústico e de linguagem. Todos os recursos construídos encontram-se publicamente disponíveis e, junto com uma interface de programação proposta, têm sido usados para o desenvolvimento de várias novas aplicações em tempo-real, incluindo um módulo de reconhecimento de voz para a suíte de aplicativos para escritório OpenOffice.org. São apresentados testes de desempenho dos sistemas desenvolvidos. Os recursos aqui produzidos e disponibilizados facilitam a adoção da tecnologia de voz para PB por outros grupos de pesquisa, desenvolvedores e pela indústria. A segunda parte do trabalho apresenta um novo método para reavaliar (rescoring) o resultado do reconhecimento baseado em HMMs, o qual é organizado em uma estrutura de dados do tipo lattice. Mais especificamente, o sistema utiliza classificadores discriminativos que buscam diminuir a confusão entre pares de fones. Para cada um desses problemas binários, são usadas técnicas de seleção automática de parâmetros para escolher a representaçãao paramétrica mais adequada para o problema em questão.

Seleção de características utilizando algoritmos evolucionistas e suas aplicações em reconhecimento de padrões

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Assessing the gain of biological data integration in gene networks inference

Background: A current challenge in gene annotation is to define the gene function in the context of the network of relationships instead of using single genes. The inference of gene networks (GNs) has emerged as an approach to better understand the biology of the system and to study how several components of this network interact with each other and keep their functions stable. However, in general there is no sufficient data to accurately recover the GNs from their expression levels leading to the curse of dimensionality, in which the number of variables is higher than samples. One way to mitigate this problem is to integrate biological data instead of using only the expression profiles in the inference process. Nowadays, the use of several biological information in inference methods had a significant increase in order to better recover the connections between genes and reduce the false positives. What makes this strategy so interesting is the possibility of confirming the known connections through the included biological data, and the possibility of discovering new relationships between genes when observed the expression data. Although several works in data integration have increased the performance of the network inference methods, the real contribution of adding each type of biological information in the obtained improvement is not clear. Methods: We propose a methodology to include biological information into an inference algorithm in order to assess its prediction gain by using biological information and expression profile together. We also evaluated and compared the gain of adding four types of biological information: (a) protein-protein interaction, (b) Rosetta stone fusion proteins, (c) KEGG and (d) KEGG+GO. Results and conclusions: This work presents a first comparison of the gain in the use of prior biological information in the inference of GNs by considering the eukaryote (P. falciparum) organism. Our results indicates that information based on direct interaction can produce a higher improvement in the gain than data about a less specific relationship as GO or KEGG. Also, as expected, the results show that the use of biological information is a very important approach for the improvement of the inference. We also compared the gain in the inference of the global network and only the hubs. The results indicates that the use of biological information can improve the identification of the most connected proteins.

Signal propagation in Bayesian networks and its relationship with intrinsically multivariate predictive variables

A set of predictor variables is said to be intrinsically multivariate predictive (IMP) for a target variable if all properly contained subsets of the predictor set are poor predictors of the. target but the full set predicts the target with great accuracy. In a previous article, the main properties of IMP Boolean variables have been analytically described, including the introduction of the IMP score, a metric based on the coefficient of determination (CoD) as a measure of predictiveness with respect to the target variable. It was shown that the IMP score depends on four main properties: logic of connection, predictive power, covariance between predictors and marginal predictor probabilities (biases). This paper extends that work to a broader context, in an attempt to characterize properties of discrete Bayesian networks that contribute to the presence of variables (network nodes) with high IMP scores. We have found that there is a relationship between the IMP score of a node and its territory size, i.e., its position along a pathway with one source: nodes far from the source display larger IMP scores than those closer to the source, and longer pathways display larger maximum IMP scores. This appears to be a consequence of the fact that nodes with small territory have larger probability of having highly covariate predictors, which leads to smaller IMP scores. In addition, a larger number of XOR and NXOR predictive logic relationships has positive influence over the maximum IMP score found in the pathway. This work presents analytical results based on a simple structure network and an analysis involving random networks constructed by computational simulations. Finally, results from a real Bayesian network application are provided. (C) 2012 Elsevier Inc. All rights reserved.

EEG amplitude modulation analysis for semi-automated diagnosis of Alzheimer's disease

Recent experimental evidence has suggested a neuromodulatory deficit in Alzheimer's disease (AD). In this paper, we present a new electroencephalogram (EEG) based metric to quantitatively characterize neuromodulatory activity. More specifically, the short-term EEG amplitude modulation rate-of-change (i.e., modulation frequency) is computed for five EEG subband signals. To test the performance of the proposed metric, a classification task was performed on a database of 32 participants partitioned into three groups of approximately equal size: healthy controls, patients diagnosed with mild AD, and those with moderate-to-severe AD. To gauge the benefits of the proposed metric, performance results were compared with those obtained using EEG spectral peak parameters which were recently shown to outperform other conventional EEG measures. Using a simple feature selection algorithm based on area-under-the-curve maximization and a support vector machine classifier, the proposed parameters resulted in accuracy gains, relative to spectral peak parameters, of 21.3% when discriminating between the three groups and by 50% when mild and moderate-to-severe groups were merged into one. The preliminary findings reported herein provide promising insights that automated tools may be developed to assist physicians in very early diagnosis of AD as well as provide researchers with a tool to automatically characterize cross-frequency interactions and their changes with disease.

Can neuroimaging be used as a support to diagnosis of borderline personality disorder? An approach based on computational neuroanatomy and machine learning

Several recent studies in literature have identified brain morphological alterations associated to Borderline Personality Disorder (BPD) patients. These findings are reported by studies based on voxel-based-morphometry analysis of structural MRI data, comparing mean gray-matter concentration between groups of BPD patients and healthy controls. On the other hand, mean differences between groups are not informative about the discriminative value of neuroimaging data to predict the group of individual subjects. In this paper, we go beyond mean differences analyses, and explore to what extent individual BPD patients can be differentiated from controls (25 subjects in each group), using a combination of automated-morphometric tools for regional cortical thickness/volumetric estimation and Support Vector Machine classifier. The approach included a feature selection step in order to identify the regions containing most discriminative information. The accuracy of this classifier was evaluated using the leave-one-subject-out procedure. The brain regions indicated as containing relevant information to discriminate groups were the orbitofrontal, rostral anterior cingulate, posterior cingulate, middle temporal cortices, among others. These areas, which are distinctively involved in emotional and affect regulation of BPD patients, were the most informative regions to achieve both sensitivity and specificity values of 80% in SVM classification. The findings suggest that this new methodology can add clinical and potential diagnostic value to neuroimaging of psychiatric disorders. (C) 2012 Elsevier Ltd. All rights reserved.

A system for classification of time-series data from industrial non-destructive device

This work proposes a system for classification of industrial steel pieces by means of magnetic nondestructive device. The proposed classification system presents two main stages, online system stage and off-line system stage. In online stage, the system classifies inputs and saves misclassification information in order to perform posterior analyses. In the off-line optimization stage, the topology of a Probabilistic Neural Network is optimized by a Feature Selection algorithm combined with the Probabilistic Neural Network to increase the classification rate. The proposed Feature Selection algorithm searches for the signal spectrogram by combining three basic elements: a Sequential Forward Selection algorithm, a Feature Cluster Grow algorithm with classification rate gradient analysis and a Sequential Backward Selection. Also, a trash-data recycling algorithm is proposed to obtain the optimal feedback samples selected from the misclassified ones.

Data Mining in Clinical Practice for the Quantification of Motor Impairment in Parkinson's Disease

Advances in biomedical signal acquisition systems for motion analysis have led to lowcost and ubiquitous wearable sensors which can be used to record movement data in different settings. This implies the potential availability of large amounts of quantitative data. It is then crucial to identify and to extract the information of clinical relevance from the large amount of available data. This quantitative and objective information can be an important aid for clinical decision making. Data mining is the process of discovering such information in databases through data processing, selection of informative data, and identification of relevant patterns. The databases considered in this thesis store motion data from wearable sensors (specifically accelerometers) and clinical information (clinical data, scores, tests). The main goal of this thesis is to develop data mining tools which can provide quantitative information to the clinician in the field of movement disorders. This thesis will focus on motor impairment in Parkinson's disease (PD). Different databases related to Parkinson subjects in different stages of the disease were considered for this thesis. Each database is characterized by the data recorded during a specific motor task performed by different groups of subjects. The data mining techniques that were used in this thesis are feature selection (a technique which was used to find relevant information and to discard useless or redundant data), classification, clustering, and regression. The aims were to identify high risk subjects for PD, characterize the differences between early PD subjects and healthy ones, characterize PD subtypes and automatically assess the severity of symptoms in the home setting.