Abstract
Machine Learning techniques have been largely applied to the problem of class prediction in microarray data. Nevertheless, current approaches to select appropriate methods for such task often result unsatisfactory in many ways, instigating the need for the development of tools to automate the process. In this context, the authors introduce the use of metalearning in the specific domain of gene expression classification. Experiments with the KNN-ranking method for algorithm recommendation applied for 49 datasets yielded successful results. © 2008 IEEE.

Abstract
Maintenance management is a subject that, instead of reducing importance, with the increase of equipment reliability, it increases its role in the companies and obliges the increase of the level of demand of professionals involved because of the new technical and environmental demands. Sometimes, scientific developments anticipate the company's needs while other times it is the company that challenges science. The maintenance area is an example that offers challenges to both science and companies in order to optimize the performance of equipment and facilities. This is also the case of wind generators, because their expansion, evolution, maintenance and reliability guarantee, needs to be adequately articulated in order to maximize production time and, obviously, to optimize maintenance interventions. It is because of this kind of challenge that the authors are developing new methodologies in the area of wind generators that aims to optimize the cycles of production and, consequently, reduce other kinds of energy production. The new features include on-line measures and the corresponding on-time treatment, using algorithms based on time-series forecasting and wireless technology to transmit the signals. The prediction models uses regression techniques based on SVR, ARMA and ARIMA models, modified according to this specific case. The weather, conditions and the technical and construction characteristics of wind generators are only some variables that we have in account in the models that are under development. But, if these conditions are important, it is also very important to collect, read and treat data from sensors placed in wind generators that, because their geographic dispersion, and difficulty of transmission, must be solved adequately and conjugated with the above referred algorithms, in order to implement an adequate system. This is the ambit of the present article that reports a wide approach of a subject that usually is managed separately, this is, the hardware from one side and the prediction algorithms from other side. This is possible because the team has being researching and developing algorithms and an information system, since many years ago, around the terology subject that is a wider vision of maintenance. Then, the new methodologies, above mentioned, will be, later, incorporated through new predictive maintenance modules in an integrated maintenance management system called SMIT (Terology Integrated Modular System). The base of SMIT is accessed through a client-server system and a browser system that includes the main modules of a traditional system, as well as a fault diagnosis module, a non-periodic maintenance planning module and a generic oncondition maintenance module, among other innovations.

Abstract
The performance of Artificial Neural Networks is largely influenced by the value of their parameters. Among these free parameters, one can mention those related with the network architecture, e.g., number of hidden neurons, number of hidden layers, activation function, and those associated with a learning algorithm, e.g., learning rate. Optimization techniques, often Genetic Algorithms, have been used to tune neural networks parameter values. Lately, other techniques inspired in Biology have been investigated. In this paper, we compare the influence of different bio-inspired optimization techniques on the accuracy obtained by the networks in the domain of gene expression analysis. The experimental results show the potential of use this techniques for parameter tuning of neural networks. © 2008 IEEE.

Abstract
Research in the RoboCup domain has grown considerably since the beginning of this initiative more than ten years ago. Much of this growth is due to the existence of different leagues, that allow the focussing of research in specific and heterogeneous issues. This specialisation of research has, though, proven to have some drawbacks: research subjects become very specific, and one loses the ability of properly generalising, and sharing, the obtained results. This paper presents an architecture that aims at being open, enabling the development of independent components that can easily be ported between application environments. This architecture, called Common Framework, relies on standardised interfaces, protocols and communication channels between components. Besides allowing the free association of heterogeneous components, like real and simulated back-ends, it also considerably eases the introduction of principles of redundancy and fault tolerance.

Abstract
This document describes how we are using mobile phones together with Google Earth to allow children to create multisensory geographic information in learning and participatory contexts.

Abstract
Distributed real-time system, such as factory automation systems, require that computer nodes communicate with a known and low hound on the communication delay. This can be achieved with traditional time division multiple access (TDMA). But improved flexibility and simpler upgrades are possible through the use of TDMA with slot-skipping (TDMA/SS), meaning that a slot is skipped whenever it is not used and consequently the slot after the skipped slot starts earlier. We propose a schedulahility analysis for TDMA/SS. We assume knowledge of all message streams in the system, and that each node schedules messages in its output queue according to deadline monotonic. Firstly, we present a non-exact (but fast) analysis and then, at the cost of computation time, we also present an algorithm that computes exact queuing times.