Abstract
This work proposes a hybrid genetic algorithm (GA) to address the unit commitment (UC) problem. In the UC problem, the goal is to schedule a subset of a given group of electrical power generating units and also to determine their production output in order to meet energy demands at minimum cost. In addition, the solution must satisfy a set of technological and operational constraints. The algorithm developed is a hybrid biased random key genetic algorithm (HBRKGA). It uses random keys to encode the solutions and introduces bias both in the parent selection procedure and in the crossover strategy. To intensify the search close to good solutions, the GA is hybridized with local search. Tests have been performed on benchmark large-scale power systems. The computational results demonstrate that the HBRKGA is effective and efficient. In addition, it is also shown that it improves the solutions obtained by current state-of-the-art methodologies.

Abstract
Despite being staggeringly error prone, spreadsheets are a highly flexible programming environment that is widely used in industry. In fact, spreadsheets are widely adopted for decision making, and decisions taken upon wrong (spreadsheet-based) assumptions may have serious economical impacts on businesses, among other consequences. This paper proposes a technique to automatically pinpoint potential faults in spreadsheets. It combines a catalog of spreadsheet smells that provide a first indication of a potential fault, with a generic spectrum-based fault localization strategy in order to improve (in terms of accuracy and false positive rate) on these initial results. Our technique has been implemented in a tool which helps users detecting faults. To validate the proposed technique, we consider a well-known and well-documented catalog of faulty spreadsheets. Our experiments yield two main results: we were able to distinguish between smells that can point to faulty cells from smells and those that are not capable of doing so; and we provide a technique capable of detecting a significant number of errors: two thirds of the cells labeled as faulty are in fact (documented) errors.

Abstract
We consider a system where an agent (Alice) aims at transmitting a message to a second agent (Bob) over a set of parallel channels, while keeping it secret from a third agent (Eve) by using physical layer security techniques. We assume that Alice perfectly knows the set of channels with respect to Bob, but she has only a statistical knowledge of the channels with respect to Eve. We derive bounds on the achievable outage secrecy rates, by considering coding either within each channel or across all parallel channels. Transmit power is adapted to the channel conditions, with a constraint on the average power over the whole transmission. We also focus on the maximum cumulative outage secrecy rate that can be achieved. Moreover, in order to assess the performance in a real life scenario, we consider the use of practical error correcting codes. We extend the definitions of security gap and equivocation rate, previously applied to the single additive white Gaussian noise channel, to Rayleigh distributed parallel channels, on the basis of the error rate targets and the outage probability. Bounds on these metrics are also derived, considering the statistics of the parallel channels. Numerical results are provided, that confirm the feasibility of the considered physical layer security techniques.

Abstract
Machine learning studies automatic methods for acquisition of domain knowledge with the goal of improving systems performance as the result of experience. In the past two decades, machine learning research and practice has focused on batch learning usually with small data sets. The rationale behind this practice is that examples are generated at random accordingly to some stationary probability distribution. Most learners use a greedy, hill-climbing search in the space of models. They are prone to overfitting, local maximas, etc. Data are scarce and statistic estimates have high variance. A paradigmatic example is the TDIT algorithm to learn decision trees [14]. As the tree grows, less and fewer examples are available to compute the sufficient statistics, variance increase leading to model instability Moreover, the growing process re-uses the same data, exacerbating the overfitting problem. Regularization and pruning mechanisms are mandatory. © Springer International Publishing Switzerland 2015.

Abstract
There is relatively little guidance on the situated ergonomic assessment of medical devices, and few case studies that detail this type of evaluation. This paper reports results of a detailed case study that focuses on the design and use of a modern blood glucose meter on an oncology ward. We spent approximately 150 h in-situ, over 11 days and 4 nights, performing observations and interviews with users. This was complemented by interviews with two staff with oversight and management responsibility related to the device. We identified 19 issues with the design and use of this device. These issues were grouped into 7 themes which can help guide the situated study of medical devices: usability, knowledge gaps and mental models, workarounds, wider tasks and equipment, the patient, connection between services, and policy. © 2014 Elsevier Ltd and The Ergonomics Society.

Abstract
Enterprise Service Bus (ESB) is a middleware infrastructure that provides a way to integrate loosely-coupled heterogeneous software applications based on the services principles. The life-cycle management of services in such environments is a critical issue for the component's reuse, maintenance and operation. This paper introduces a service life-cycle management module that extends the traditional functionalities with advanced monitoring and data analytics to contribute for the robustness, reliability and self-organization of networks of clusters based on ESB platforms. The realization of this module was embedded in the JBoss ESB, considering a sniffer mechanism to collect the service messages crossing the bus and a Liferay portal to display relevant information related to the services' health.