Francisco Madeiro

Hybrid intelligent system for air quality forecasting using phase adjustment

Abstract The pollution caused by particulate matter (PM) concentration has a negative impact on population health, due to its relationship with several diseases. In this sense, several intelligent systems have been proposed for forecasting the PM concentration. Although it is known in the literature that PM concentration time series behave like random walk, to the authors’ knowledge there is no intelligent systems developed to forecast the PM concentration that consider this characteristic. In this paper, we present an architecture developed to forecast time series guided by random walk process. The architecture, called Time-delay Added Evolutionary Forecasting (TAEF), consists of two steps: parameters optimization and phase adjustment. In the first step, a genetic optimization procedure is employed to adjust the parameters of a Multilayer Perceptron neural network that is used as the prediction model. The genetic algorithm adjusts the following parameters of the prediction model: the number of input nodes (time lags), the number of neurons in the hidden layer and the training algorithm. The second step is performed aiming to reduce the difference between the forecasting and the actual concentration value of the time series, that occur in the forecasting of the time series with random walk behavior. The approach is data-driven and only uses the past values of the pollutant concentrations to predict the next day concentration; in other words, it does not require any exogenous information. The experimental study is performed using time series of concentration levels of particulate matter (PM2.5 and PM10) from Helsinki and shows that the approach overcomes previous state-of-the-art methods by a large margin.

Time-series forecasting of pollutant concentration levels using particle swarm optimization and artificial neural networks

This study evaluates the application of an intelligent hybrid system for time-series forecasting of atmospheric pollutant concentration levels. The proposed method consists of an artificial neural network combined with a particle swarm optimization algorithm. The method not only searches relevant time lags for the correct characterization of the time series, but also determines the best neural network architecture. An experimental analysis is performed using four real time series and the results are shown in terms of six performance measures. The experimental results demonstrate that the proposed methodology achieves a fair prediction of the presented pollutant time series by using compact networks.

An Evolutionary Approach for Vector Quantization Codebook Optimization

This paper proposes a hybrid evolutionary algorithm based on an accelerated version of K-means integrated with a modified genetic algorithm (GA) for vector quantization (VQ) codebook optimization. From simulation results involving image compression based on VQ, it is observed that the proposed method leads to better codebooks when compared with the conventional one (GA + standard K-means), in the sense that the former leads to higher peak signal-to-noise ratio (PSNR) results for the reconstructed images. Additionally, it is observed that the proposed method requires fewer GA generations (up to 40%) to achieve the best PSNR results produced by the conventional method.

Tecnologias da Informação e da Comunicação (TIC) aplicadas à dislexia: revisão de literatura

O objetivo deste estudo e apresentar uma revisao integrativa da literatura, contemplando artigos cientificos publicados em periodicos nacionais e internacionais que abordam o uso das tecnologias da informacao e da comunicacao (TIC), tais como computador, tablets, iPads, mobile phones, e-readers, realidade virtual e ambiente virtual de aprendizagem, aplicadas a dislexia. A base de dados escolhida para este estudo foi constituida de artigos cientificos publicados no periodo de 2010 a 2015, a partir das seguintes bases eletronicas de dados: Science Direct/Elsevier, SciELO - Scientific Electronic Library Online, MedLine - Medical Literature Analysis and Retrieval e o Portal de Periodicos da CAPES. Foram selecionados para este estudo 21 artigos cientificos, sendo 20 (95,23%) artigos internacionais e um (4,77%) artigo nacional. Os trabalhos contemplados, no presente estudo, em geral, visam a construcao e aplicacao de instrumentos tecnologicos que possam vir a minimizar as dificuldades do dislexico no âmbito da aprendizagem da leitura e da escrita. Em meio a escassez de artigos publicados no Brasil, verifica-se a necessidade de mais estudos sobre essa tematica, tendo em vista os beneficios das TIC no âmbito da avaliacao e intervencao em dislexia constatados em artigos internacionais.

Multiresolution codebook design for wavelet/VQ image coding

Image compression using the combination of the discrete wavelet transform (DWT) with vector quantization (VQ) has been considered in many works. Most of the studies have been dedicated to evaluating the choice of the wavelet filters employed in the multiresolution wavelet image decomposition or to developing bit-allocation schemes. It is worth mentioning, however, that the VQ codebook design plays a crucial role in the quality of the reconstructed image. A competitive neural network algorithm (synaptic space competitive algorithm, SSC) has already been successfully applied for voice waveform VQ codebook design (Vilar Franca and Aguiar Neto, 1994). In the present work, the SSC algorithm is applied in a wavelet/VQ image coding framework. The SSC codebooks are used to code the image subbands that result from the multiresolution decomposition. The coding results show that the SSC multiresolution codebooks lead to better reconstructed image quality than that obtained by using JPEG and conventional (spatial domain) VQ.

An Approach to Improve the Performance of PM Forecasters

The particulate matter (PM) concentration has been one of the most relevant environmental concerns in recent decades due to its prejudicial effects on living beings and the earth’s atmosphere. High PM concentration affects the human health in several ways leading to short and long term diseases. Thus, forecasting systems have been developed to support decisions of the organizations and governments to alert the population. Forecasting systems based on Artificial Neural Networks (ANNs) have been highlighted in the literature due to their performances. In general, three ANN-based approaches have been found for this task: ANN trained via learning algorithms, hybrid systems that combine search algorithms with ANNs, and hybrid systems that combine ANN with other forecasters. Independent of the approach, it is common to suppose that the residuals (error series), obtained from the difference between actual series and forecasting, have a white noise behavior. However, it is possible that this assumption is infringed due to: misspecification of the forecasting model, complexity of the time series or temporal patterns of the phenomenon not captured by the forecaster. This paper proposes an approach to improve the performance of PM forecasters from residuals modeling. The approach analyzes the remaining residuals recursively in search of temporal patterns. At each iteration, if there are temporal patterns in the residuals, the approach generates the forecasting of the residuals in order to improve the forecasting of the PM time series. The proposed approach can be used with either only one forecaster or by combining two or more forecasting models. In this study, the approach is used to improve the performance of a hybrid system (HS) composed by genetic algorithm (GA) and ANN from residuals modeling performed by two methods, namely, ANN and own hybrid system. Experiments were performed for PM2.5 and PM10 concentration series in Kallio and Vallila stations in Helsinki and evaluated from six metrics. Experimental results show that the proposed approach improves the accuracy of the forecasting method in terms of fitness function for all cases, when compared with the method without correction. The correction via HS obtained a superior performance, reaching the best results in terms of fitness function and in five out of six metrics. These results also were found when a sensitivity analysis was performed varying the proportions of the sets of training, validation and test. The proposed approach reached consistent results when compared with the forecasting method without correction, showing that it can be an interesting tool for correction of PM forecasters.

Terrain-Based Memetic Algorithms for Vector Quantizer Design

Recently, a Genetic Accelerated K-Means Algorithm (GAKM) was proposed as an approach for optimizing Vector Quantization (VQ) codebooks, relying on an accelerated version of K-Means algorithm as a new local learning module. This approach requires the determination of a scale factor parameter (η), which affects the local search performed by GAKM. The problem of auto-adapting the local search in GAKM, by adjusting the η parameter, is addressed in this work by the proposal of a Terrain-Based Memetic Algorithm (TBMA), derived from existing spatially distributed evolutionary models. Simulation results regarding image VQ show that this new approach is able to adjust the scale factor (η) for different images at distinct coding rates, leading to better Peak Signal-to-Noise Ratio values for the reconstructed images when compared to both K-Means and Cellular Genetic Algorithm + K-Means. The TBMA also demonstrates capability of tuning the mutation rate throughout the genetic search.

Improving Image Vector Quantization with a Genetic Accelerated K-Means Algorithm

In this paper, vector quantizer optimization is accomplished by a hybrid evolutionary method, which consists of a modified genetic algorithm (GA) with a local optimization module given by an accelerated version of the K -means algorithm. Simulation results regarding image compression based on VQ show that the codebooks optimized by the proposed method lead to reconstructed images with higher peak signal-to-noise ratio (PSNR) values and that the proposed method requires fewer GA generations (up to 40%) to achieve the best PSNR results produced by the conventional GA + standard K -means approach. The effect of increasing the number of iterations performed by the local optimization module within the proposed method is discussed.

A Fish School Search based algorithm for image Channel-Optimized Vector Quantization

Channel-Optimized Vector Quantization (COVQ) is an alternative to Vector Quantization (VQ) in the scenario of transmission over noisy channels. The codebook design is an optimization problem in which a set of vectors must be optimized to represent the signals to be quantized. This paper presents a new approach to COVQ codebook design, which is a challenging optimization problem. The proposed technique embeds the Fish School Search (FSS) as a Swarm Clustering Algorithm to COVQ. Simulation results concerning a Binary Symmetric Channel (BSC) reveal the superiority of the proposed technique over conventional COVQ codebook design in terms of the quality of reconstructed images.

Algoritmo do Vagalume Aplicado ao Projeto de Dicionários do COVQ

Resumo—Este artigo apresenta um método de aplicação do algoritmo do vagalume à etapa de projeto de dicionários do quantizador vetorial otimizado para canal, projetando dicionários para a quantização vetorial de imagens com transmissão por canal binário simétrico. O método, chamado de FA-COVQ (Firefly Algorithm – Channel Optimized Vector Quantization), buscou otimizar o valor de distorção média gerado no projeto de dicionários do quantizador vetorial otimizado para canal e, com isso, produzir dicionários mais robustos aos erros de canal, obtendo imagens reconstruı́das de melhor qualidade. Os resultados mostram que o FA-COVQ reduziu em 93,61% dos casos a distorção média final dos dicionários projetados, quando comparado ao quantizador vetorial otimizado para canal. Este método também conseguiu melhorar o valor de relação sinalruı́do de pico das imagens reconstruı́das: 91,66% dos conjuntos de dicionários do FA-COVQ avaliados no estudo apresentaram imagens reconstruı́das com valor de relação sinal-ruı́do de pico maiores que as imagens reconstruı́das pelos conjuntos de dicionários do quantizador vetorial otimizado para canal, sendo que este número sobe para 100,00% dos casos, quando o canal binário simétrico apresenta probabilidade de erro de bit de 0,005, de 0,010 e de 0,050.