Abstract

Abstract
Integration methods for ensemble learning can use two different approaches: combination or selection. The combination approach (also called fusion) consists on the combination of the predictions obtained by different models in the ensemble to obtain the final ensemble predication. The selection approach selects one (or more) models from the ensemble according to the prediction performance of these models on similar data from the validation set. Usually, the method to select similar data is the k-nearest neighbors with the Euclidean distance. In this paper we discuss other approaches to obtain similar data for the regression problem. We show that using similarity measures according to the target values improves results. We also show that selecting dynamically several models for the prediction task increases prediction accuracy comparing to the selection of just one model.

Abstract
Travel time prediction is an important tool for the planning tasks of mass transit and logistics companies. ID this paper we investigate the use of regression methods for the problem of predicting the travel time of buses in a Portuguese public transportation company. More specifically, we empirically evaluate the impact of varying parameters on the performance of different regression algorithms, such as support vector machines (SVM), random forests (RF) and projection pursuit, regression (PPR). We also evaluate the impact of the focusing tusks (example selection; domain value definition and feature selection) in the accuracy of those algorithms. Concerning the algorithms, we observe that 1) RF is quite robust to the choice of parameters and focusing methods: 2) the choice of parameters for SVM can be made independently of focusing methods while 3) for PPR they should be selected simultaneously. For the focusing methods, we observe that a stronger effect is obtained using example selection, particularly in combination with SVM.

Abstract
The purpose of this research is to develop a process capable of optimizing the capture of light in a circumscribed space, using a responsive system of skylights. Research is focused on the design of irregular dome-like pavilion spaces circumscribed by curved surfaces, but the envisioned process might be applied to other functional, formal, and spatial typologies. The design of the pavilion starts with the generation of a surface, later tessellated into a matrix of skylights depending on the geographic location and sun orientation. In the constructed pavilion, the skylights react to the variation of daylight intensity during the day to satisfy specified internal lighting needs. Simultaneously, the integration of conception and fabrication using digital tools (CAD-CAM) facilitates the construction of a non-standard, parametric geometry, thereby diminishing the costs of production and allowing for personalization, while assuring global sustainability.

Abstract
In this text is proposed a simple method for estimating or characterize the behaviour of complex systems, in particular when these are being studied throughout simulation. Usual ways of treating the complex output data obtained from the activity (real or simulated) of such a kind of systems, which in many cases people classify and analyse along the time domain, usually the most, complex perspective, is herein substituted by the idea of representing such data in the frequency domain, somehow like what is commonly done in Fourier Analysis and in Quantum Mechanics. This is expected to give the analyst a more holistic perspective oil the system's behaviour, as well as letting him/her choose almost freely the complex states in which such behaviour is to be projected. We hope this will lead to simpler processes in characterizing complex systems.

Abstract
This paper is divided into four parts: the first one introduces SADEX, a fuzzy Case Based Reasoning (CBR) System for fault diagnosis. The second focus on its observation relevance factors and shows how the results are in complete agreement with the relevance concept introduced by Robertson and Spark-Jones in their well known and proved technique for document retrieval. The third describes how equipment composition information can be used to generalize and adapt case solutions to new and unknown occurrences; this generalization is based on a taxonomic similarity between functionally autonomous modules (FAMs). Finally the MKM-Maintenance Knowledge Manager system is introduced.