Abstract
The class imbalance problem has been thoroughly studied over the past two decades. More recently, the research community realized that the problem of imbalanced distributions also occurred in other tasks beyond classification. Regression problems are among these newly studied tasks where the problem of imbalanced domains also poses important challenges. Imbalanced regression problems occur in a diversity of real world domains such as meteorological (predicting weather extreme values), financial (extreme stock returns forecasting) or medical (anticipate rare values). In imbalanced regression the end-user preferences are biased towards values of the target variable that are under-represented on the available data. Several pre-processing methods were proposed to address this problem. These methods change the training set to force the learner to focus on the rare cases. However, as far as we know, the relationship between the data intrinsic characteristics and the performance achieved by these methods has not yet been studied for imbalanced regression tasks. In this paper we describe a study of the impact certain data characteristics may have in the results of applying pre-processing methods to imbalanced regression problems. To achieve this goal, we define potentially interesting data characteristics of regression problems. We then conduct our study using a synthetic data repository build for this purpose. We show that all the different characteristics studied have a different behaviour that is related with the level at which the data characteristic is present and the learning algorithm used. The main contributions of our work are: i) to define interesting data characteristics for regression tasks; ii) to create the first repository of imbalanced regression tasks containing 6000 data sets with controlled data characteristics; and iii) to provide insights on the impact of intrinsic data characteristics in the results of pre-processing methods for handling imbalanced regression tasks.

Abstract
In this paper, we describe a simple and indirect method to evaluate the kinetics of the optical properties for biological tissues under optical clearing treatments. We use the theoretical formalism in this method to process experimental data obtained from colorectal muscle samples to evaluate and characterize the dehydration and refractive index matching mechanisms.

Abstract
Interoperability is a challenge for the realisation of smart grids. In this work, we apply the methodology for interoperability testing and the design of experiments developed at the Smart Grids Interoperability Laboratory of the Joint Research Centre of the European Commission on a simple use case. The methodology is based on the Smart Grid Architecture Model (SGAM) of CEN/CENELEC/ETSI and includes the concept of Basic Application Profiles (BAP) and Basic Application Interoperability Profiles (BAIOP). The relevant elements of the methodology are the design of experiments and the sensitivity/uncertainty analysis, which can reveal the limits of a system under test and give valuable feedback about the critical conditions which do not guarantee interoperability. The design and analysis of experiments employed in the Joint Research Centre (JRC) methodology supply information about the crucial parameters that either lead to an acceptable system performance or to a failure of interoperability. The use case on which the methodology is applied describes the interaction between a data concentrator and one or more smart meters. Experimental results are presented that show the applicability of the methodology and the design of experiments in practice. The system is tested under different conditions by varying two parameters: the rate at which meter data are requested by the data concentrator and the number of smart meters connected to the data concentrator. With this use case example the JRC methodology is illustrated at work, and its effectiveness for testing interoperability of a system under stress conditions is highlighted.

Abstract
Virtual reality (VR) games have the potential to produce immersive experiences. To better explore the potential of VR games, it becomes necessary to understand what affects the player's presence in VR games. This work measures and compares the levels of presence and cybersickness in VR environments. Two games with different levels of interaction and narrative were compared. Presence and cybersickness were measured in a sample of 32 subjects using the IPQp questionnaire and a Portuguese version of the SSQ respectively. The results indicate that there were no differences in presence and cybersickness between the interaction and the narrative dimensions. To extend the study, the gender of participants was also considered an independent variable where we found significant differences in the metrics of presence and experienced realism, nausea and disorientation with female participants getting higher scores. © 2018 IEEE.

Abstract
To counter the intermittent nature of variable Renewable Energy Sources (vRESs), it is necessary to deploy new technologies that increase the flexibility dimension in distribution systems. In this framework, the current work presents an extensive analysis on the level of energy storage systems (ESSs) in order to add flexibility, and handle the intermittent nature of vRES. Moreover, this work provides an operational model to optimally manage a distribution system that encompasses large quantities of vRESs by means of ESSs. The model is of a stochastic mixed integer linear programming (WILY) nature, which uses a linearized AC optimal power flow network model. The standard IEEE 119-bus test system is used as a case study. Generally, numerical results show that ESSs enable a much bigger portion of the final energy consumption to be met by vRES power, generated locally.

Abstract
Mobile robot applications are increasing its usability in industries and services (examples: vacuum cleaning, painting and farming robots, among others). Some of the applications require that the robot moves in an environment between two positions while others require that the robot scans all the positions (Coverage Path Planning). Optimizing the traveled distance or the time to scan the path, should be done in order to reduce the costs. This paper addresses an optimization approach of the coverage path planning using Q-Learning algorithm. Comparison with other methods allows to validate the methodology.