Abstract
Photovoltaic (PV) solar power capacity is growing in several countries, either concentrated in medium/large size solar parks or distributed by the medium and low voltage grid. Solar power forecasting is a key input for supporting grid management, participation in the electricity market and maintenance planning. This paper proposes a new forecasting system that is a hybrid of different models, such as electrical and statistical models, as well as different Numerical Weather Prediction (NWP) sources. The time horizon is 48 hours ahead. The proposed model was operationalized and tested in a real PV installation located in North Portugal with 16 kW.

Abstract
Chord progressions are widely used in music. The automatic generation of chord progressions can be challenging because it depends on many factors, such as the musical context, personal preference, and aesthetic choices. In this work, we propose a penalty function that encodes musical rules to automatically generate chord progressions. Then we use an artificial immune system (AIS) to minimize the penalty function when proposing candidates for the next chord in a sequence. The AIS is capable of finding multiple optima in parallel, resulting in several different chords as appropriate candidates. We performed a listening test to evaluate the chords subjectively and validate the penalty function. We found that chords with a low penalty value were considered better candidates than chords with higher penalty values.

Abstract
The use of unmanned vehicle networks for diverse applications is becoming widespread. It is generally hard to program unmanned vehicle networks as a "whole", however. The coordination of multiple vehicles requires careful planning through intricate human intervention, and a high degree of informality is implied in what concerns the specification of a "network program" for an application scenario. In this context, we have been developing a programming language for expressing global specifications of coordinated behavior in unmanned vehicle networks, the Networked Vehicles' Language (NVL). In this paper we illustrate the use of the language for a thermal pollution plume tracking scenario employing unmanned underwater vehicles.

Abstract
Distributed Generation (DG) is providing end consumers the possibility to satisfy part of their electricity consumption by using their own small-scale power generators. To regulate DG, new regulation schemes are needed, being net metering one of the most used. However, regulators appreciate that net metering could jeopardize the incomes to cover the regulated activities' cost. This paper proposes a mathematical bi-level model to obtain the evolution of the access tariffs and their corresponding incomes needed to cover the regulated costs, as well as the optimal DG investment of the consumers under a net metering regulation, in a simplified framework. © 2015 IEEE.

Abstract
Adverse Drug Events (ADEs) are a major health problem, and developing accurate prediction methods may have a significant impact in public health. Ideally, we would like to have predictive methods, that could pinpoint possible ADRs during the drug development process. Unfortunately, most relevant information on possible ADRs is only available after the drug is commercially available. As a first step, we propose using prior information on existing interactions through recommendation systems algorithms. We have evaluated our proposal using data from the ADReCS database with promising results.

Abstract
Probabilistic Inductive Logic Programming (PILP) is a relatively unexplored area of Statistical Relational Learning which extends classic Inductive Logic Programming (ILP). Within this scope, we introduce SkILL, a Stochastic Inductive Logic Learner, which takes probabilistic annotated data and produces First Order Logic (FOL) theories. Data in several domains such as medicine and bioinformatics have an inherent degree of uncertainty, and because SkILL can handle this type of data, the models produced for these areas are closer to reality. SkILL can then use probabilistic data to extract non-trivial knowledge from databases, and also address efficiency issues by introducing an efficient search strategy for finding hypotheses in PILP environments. SkILL's capabilities are demonstrated using a real world medical dataset in the breast cancer domain.