PDFOS: PDF estimation based over-sampling for imbalanced two-class problems

This contribution proposes a novel probability density function (PDF) estimation based over-sampling (PDFOS) approach for two-class imbalanced classification problems. The classical Parzen-window kernel function is adopted to estimate the PDF of the positive class. Then according to the estimated PDF, synthetic instances are generated as the additional training data. The essential concept is to re-balance the class distribution of the original imbalanced data set under the principle that synthetic data sample follows the same statistical properties. Based on the over-sampled training data, the radial basis function (RBF) classifier is constructed by applying the orthogonal forward selection procedure, in which the classifier’s structure and the parameters of RBF kernels are determined using a particle swarm optimisation algorithm based on the criterion of minimising the leave-one-out misclassification rate. The effectiveness of the proposed PDFOS approach is demonstrated by the empirical study on several imbalanced data sets.

Nonlinear identification using orthogonal forward regression with nested optimal regularization

An efficient data based-modeling algorithm for nonlinear system identification is introduced for radial basis function (RBF) neural networks with the aim of maximizing generalization capability based on the concept of leave-one-out (LOO) cross validation. Each of the RBF kernels has its own kernel width parameter and the basic idea is to optimize the multiple pairs of regularization parameters and kernel widths, each of which is associated with a kernel, one at a time within the orthogonal forward regression (OFR) procedure. Thus, each OFR step consists of one model term selection based on the LOO mean square error (LOOMSE), followed by the optimization of the associated kernel width and regularization parameter, also based on the LOOMSE. Since like our previous state-of-the-art local regularization assisted orthogonal least squares (LROLS) algorithm, the same LOOMSE is adopted for model selection, our proposed new OFR algorithm is also capable of producing a very sparse RBF model with excellent generalization performance. Unlike our previous LROLS algorithm which requires an additional iterative loop to optimize the regularization parameters as well as an additional procedure to optimize the kernel width, the proposed new OFR algorithm optimizes both the kernel widths and regularization parameters within the single OFR procedure, and consequently the required computational complexity is dramatically reduced. Nonlinear system identification examples are included to demonstrate the effectiveness of this new approach in comparison to the well-known approaches of support vector machine and least absolute shrinkage and selection operator as well as the LROLS algorithm.

A fast adaptive tunable RBF network for nonstationary systems

This paper describes a novel on-line learning approach for radial basis function (RBF) neural network. Based on an RBF network with individually tunable nodes and a fixed small model size, the weight vector is adjusted using the multi-innovation recursive least square algorithm on-line. When the residual error of the RBF network becomes large despite of the weight adaptation, an insignificant node with little contribution to the overall system is replaced by a new node. Structural parameters of the new node are optimized by proposed fast algorithms in order to significantly improve the modeling performance. The proposed scheme describes a novel, flexible, and fast way for on-line system identification problems. Simulation results show that the proposed approach can significantly outperform existing ones for nonstationary systems in particular.

The Diffusion Kernel Filter

A particle filter method is presented for the discrete-time filtering problem with nonlinear ItA ` stochastic ordinary differential equations (SODE) with additive noise supposed to be analytically integrable as a function of the underlying vector-Wiener process and time. The Diffusion Kernel Filter is arrived at by a parametrization of small noise-driven state fluctuations within branches of prediction and a local use of this parametrization in the Bootstrap Filter. The method applies for small noise and short prediction steps. With explicit numerical integrators, the operations count in the Diffusion Kernel Filter is shown to be smaller than in the Bootstrap Filter whenever the initial state for the prediction step has sufficiently few moments. The established parametrization is a dual-formula for the analysis of sensitivity to gaussian-initial perturbations and the analysis of sensitivity to noise-perturbations, in deterministic models, showing in particular how the stability of a deterministic dynamics is modeled by noise on short times and how the diffusion matrix of an SODE should be modeled (i.e. defined) for a gaussian-initial deterministic problem to be cast into an SODE problem. From it, a novel definition of prediction may be proposed that coincides with the deterministic path within the branch of prediction whose information entropy at the end of the prediction step is closest to the average information entropy over all branches. Tests are made with the Lorenz-63 equations, showing good results both for the filter and the definition of prediction.

The univalent polynomial of Suffridge as a summability kernel

A positive summability trigonometric kernel {K(n)(theta)}(infinity)(n=1) is generated through a sequence of univalent polynomials constructed by Suffridge. We prove that the convolution {K(n) * f} approximates every continuous 2 pi-periodic function f with the rate omega(f, 1/n), where omega(f, delta) denotes the modulus of continuity, and this provides a new proof of the classical Jackson`s theorem. Despite that it turns out that K(n)(theta) coincide with positive cosine polynomials generated by Fejer, our proof differs from others known in the literature.

Traffic and Road Sign Recognition

This thesis presents a system to recognise and classify road and traffic signs for the purpose of developing an inventory of them which could assist the highway engineers’ tasks of updating and maintaining them. It uses images taken by a camera from a moving vehicle. The system is based on three major stages: colour segmentation, recognition, and classification. Four colour segmentation algorithms are developed and tested. They are a shadow and highlight invariant, a dynamic threshold, a modification of de la Escalera’s algorithm and a Fuzzy colour segmentation algorithm. All algorithms are tested using hundreds of images and the shadow-highlight invariant algorithm is eventually chosen as the best performer. This is because it is immune to shadows and highlights. It is also robust as it was tested in different lighting conditions, weather conditions, and times of the day. Approximately 97% successful segmentation rate was achieved using this algorithm.Recognition of traffic signs is carried out using a fuzzy shape recogniser. Based on four shape measures - the rectangularity, triangularity, ellipticity, and octagonality, fuzzy rules were developed to determine the shape of the sign. Among these shape measures octangonality has been introduced in this research. The final decision of the recogniser is based on the combination of both the colour and shape of the sign. The recogniser was tested in a variety of testing conditions giving an overall performance of approximately 88%.Classification was undertaken using a Support Vector Machine (SVM) classifier. The classification is carried out in two stages: rim’s shape classification followed by the classification of interior of the sign. The classifier was trained and tested using binary images in addition to five different types of moments which are Geometric moments, Zernike moments, Legendre moments, Orthogonal Fourier-Mellin Moments, and Binary Haar features. The performance of the SVM was tested using different features, kernels, SVM types, SVM parameters, and moment’s orders. The average classification rate achieved is about 97%. Binary images show the best testing results followed by Legendre moments. Linear kernel gives the best testing results followed by RBF. C-SVM shows very good performance, but ?-SVM gives better results in some case.

Central limit theorem for asymmetric kernel functionals

Asymmetric kernels are quite useful for the estimation of density functions with bounded support. Gamma kernels are designed to handle density functions whose supports are bounded from one end only, whereas beta kernels are particularly convenient for the estimation of density functions with compact support. These asymmetric kernels are nonnegative and free of boundary bias. Moreover, their shape varies according to the location of the data point, thus also changing the amount of smoothing. This paper applies the central limit theorem for degenerate U-statistics to compute the limiting distribution of a class of asymmetric kernel functionals.

The aliasing effect, the Fejer Kernel and temporally aggregated long memory processes

This paper derives the spectral density function of aggregated long memory processes in light of the aliasing effect. The results are different from previous analyses in the literature and a small simulation exercise provides evidence in our favour. The main result point to that flow aggregates from long memory processes shall be less biased than stock ones, although both retain the degree of long memory. This result is illustrated with the daily US Dollar/ French Franc exchange rate series.

Avaliação de compostos com actividade inibidora da glicoproteína-P

A resistência a múltiplos fármacos é um grande problema na terapia anti-cancerígena, sendo a glicoproteína-P (P-gp) uma das responsáveis por esta resistência. A realização deste trabalho incidiu principalmente no desenvolvimento de modelos matemáticos/estatísticos e “químicos”. Para os modelos matemáticos/estatísticos utilizamos métodos de Machine Learning como o Support Vector Machine (SVM) e o Random Forest, (RF) em relação aos modelos químicos utilizou-se farmacóforos. Os métodos acima mencionados foram aplicados a diversas proteínas P-gp, p53 e complexo p53-MDM2, utilizando duas famílias: as pifitrinas para a p53 e flavonóides para P-gp e, em menor medida, um grupo diversificado de moléculas de diversas famílias químicas. Nos modelos obtidos pelo SVM quando aplicados à P-gp e à família dos flavonóides, obtivemos bons valores através do kernel Radial Basis Function (RBF), com precisão de conjunto de treino de 94% e especificidade de 96%. Quanto ao conjunto de teste com previsão de 70% e especificidade de 67%, sendo que o número de falsos negativos foi o mais baixo comparativamente aos restantes kernels. Aplicando o RF à família dos flavonóides verificou-se que o conjunto de treino apresenta 86% de precisão e uma especificidade de 90%, quanto ao conjunto de teste obtivemos uma previsão de 70% e uma especificidade de 60%, existindo a particularidade de o número de falsos negativos ser o mais baixo. Repetindo o procedimento anterior (RF) e utilizando um total de 63 descritores, os resultados apresentaram valores inferiores obtendo-se para o conjunto de treino 79% de precisão e 82% de especificidade. Aplicando o modelo ao conjunto de teste obteve-se 70% de previsão e 60% de especificidade. Comparando os dois métodos, escolhemos o método SVM com o kernel RBF como modelo que nos garante os melhores resultados de classificação. Aplicamos o método SVM à P-gp e a um conjunto de moléculas não flavonóides que são transportados pela P-gp, obteve-se bons valores através do kernel RBF, com precisão de conjunto de treino de 95% e especificidade de 93%. Quanto ao conjunto de teste, obtivemos uma previsão de 70% e uma especificidade de 69%, existindo a particularidade de o número de falsos negativos ser o mais baixo. Aplicou-se o método do farmacóforo a três alvos, sendo estes, um conjunto de inibidores flavonóides e de substratos não flavonóides para a P-gp, um grupo de piftrinas para a p53 e um conjunto diversificado de estruturas para a ligação da p53-MDM2. Em cada um dos quatro modelos de farmacóforos obtidos identificou-se três características, sendo que as características referentes ao anel aromático e ao dador de ligações de hidrogénio estão presentes em todos os modelos obtidos. Realizando o rastreio em diversas bases de dados utilizando os modelos, obtivemos hits com uma grande diversidade estrutural.

Separação cega de fontes lineares e não lineares usando algoritmo genético, redes neurais artificiais RBF e negentropia de Rényi como medida de independência

Conventional methods to solve the problem of blind source separation nonlinear, in general, using series of restrictions to obtain the solution, often leading to an imperfect separation of the original sources and high computational cost. In this paper, we propose an alternative measure of independence based on information theory and uses the tools of artificial intelligence to solve problems of blind source separation linear and nonlinear later. In the linear model applies genetic algorithms and Rényi of negentropy as a measure of independence to find a separation matrix from linear mixtures of signals using linear form of waves, audio and images. A comparison with two types of algorithms for Independent Component Analysis widespread in the literature. Subsequently, we use the same measure of independence, as the cost function in the genetic algorithm to recover source signals were mixed by nonlinear functions from an artificial neural network of radial base type. Genetic algorithms are powerful tools for global search, and therefore well suited for use in problems of blind source separation. Tests and analysis are through computer simulations

Estimates for the Poisson kernel and Hardy spaces on compact manifolds

We study Hardy spaces on the boundary of a smooth open subset or R-n and prove that they can be defined either through the intrinsic maximal function or through Poisson integrals, yielding identical spaces. This extends to any smooth open subset of R-n results already known for the unit ball. As an application, a characterization of the weak boundary values of functions that belong to holomorphic Hardy spaces is given, which implies an F. and M. Riesz type theorem. (C) 2004 Elsevier B.V. All rights reserved.

Comparative study between RBF and radial-PPS neural networks

The study of function approximation is motivated by the human limitation and inability to register and manipulate with exact precision the behavior variations of the physical nature of a phenomenon. These variations are referred to as signals or signal functions. Many real world problem can be formulated as function approximation problems and from the viewpoint of artificial neural networks these can be seen as the problem of searching for a mapping that establishes a relationship from an input space to an output space through a process of network learning. Several paradigms of artificial neural networks (ANN) exist. Here we will be investigated a comparative of the ANN study of RBF with radial Polynomial Power of Sigmoids (PPS) in function approximation problems. Radial PPS are functions generated by linear combination of powers of sigmoids functions. The main objective of this paper is to show the advantages of the use of the radial PPS functions in relationship traditional RBF, through adaptive training and ridge regression techniques.

A DOE based approach for the design of RBF artificial neural networks applied to prediction of surface roughness in AISI 52100 hardened Steel turning

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Characterization of the Pulp and Kernel Oils from Syagrus oleracea, Syagrus romanzoffiana, and Acrocomia aculeata

Vegetable oils are important sources of essential fatty acids. It is, therefore, important to characterize plant species that can be used as new oil sources. This study aimed to characterize the oils from guariroba (Syagrus oleracea), jeriva (Syagrus romanzoffiana), and macauba (Acrocomia aculeata). The physicochemical characterization was performed using official analytical methods for oils and fats, free fatty acids, peroxide value, refractive index, iodine value, saponification number, and unsaponifiable matter. The oxidative stability was determined using the Rancimat at 110 degrees C. The fatty acid composition was performed by gas chromatography. The results were submitted to Tukey's test for the medium to 5% using the ESTAT program. The pulp oils were more unsaturated than kernel oils, as evidenced by the higher refractive index and iodine value, especially the macauba pulp oil which gave 1.4556 and 80 g I(2)/100 g, respectively, for these indices. The kernel oils were less altered by oxidative process and had high induction period, free fatty acids below 0.5%, and peroxide value around 0.19 meq/kg. The guariroba kernel oil showed the largest induction period, 91.82 h.