Abstract
In this study the case of the energy consumption profile of the Aristotle University of Thessaloniki, in Greece, is presented and statistically analyzed by clustering methods on the basis of seasonal daily load curves and load shape factors, using data from real-time measurements. The results indicate that the categorization of active power demand in university buildings is an extremely useful tool for understanding and predicting the seasonal, hourly and daily energy consumption changes, which is the first step towards adopting energy efficiency policies in such scale premises as well as performing demand-side actions aiming to achieve a more economical and environmentally sustainable energy usage.

Abstract
The main goal of recommender systems is to assist users in finding items of their interest in very large collections. The use of good automatic recommendation promotes customer loyalty and user satisfaction because it helps users to attain their goals. Current methods focus on the immediate value of recommendations and are evaluated as such. This is insufficient for long term goals, either defined by users or by platform managers. This is of interest in recommending learning resources to learn a target concept, and also when a company is organizing a campaign to lead users to buy certain products or moving to a different customer segment. Therefore, we believe that it would be useful to develop recommendation algorithms that promote the goals of users and platform managers (e.g. e-shop manager, e-learning tutor, ministry of culture promotor). Accordingly, we must define appropriate evaluation methodologies and demonstrate the concept on practical cases.

Abstract
Application frameworks are a powerful technique for large-scale reuse but often very hard to learn from scratch. Although good documentation helps on reducing the learning curve, it is often found lacking, and costly, as it needs to attend different audiences with disparate learning needs. When code and documentation prove insufficient, developers turn to their network of experts. The lack of awareness about the experts, interrupting the wrong people, and experts unavailability are well known hindrances to effective collaboration. This paper presents the DRIVER platform, a collaborative learning environment for framework users to share their knowledge. It provides the documentation on a wiki, where the learning paths of the community of learners can be captured, shared, rated, and recommended, thus tapping into the collective knowledge of the community of framework users. The tool can be obtained at http://bit.ly/driverTool.

Abstract
Reject option is a technique used to improve classifier’s reliability in decision support systems. It consists in withholding the automatic classification of an item, if the decision is considered not sufficiently reliable. The rejected item is then handled by a different classifier or by a human expert. The vast majority of the works on this issue has been concerned with the development of reject option mechanisms to be used by supervised learning architectures (e.g., MLP, LVQ or SVM). In this paper, however, we aim at proposing alternatives to this view, which are based on the self-organizing map (SOM), originally an unsupervised learning scheme, but that has also been successfully used in the design of prototype-based classifiers. The basic hypothesis we defend is that it is possible to design SOM-based classifiers endowed with reject option mechanisms whose performances are comparable to or better than those achieved by standard supervised classifiers. For this purpose, we carried out a comprehensively evaluation of the proposed SOM-based classifiers on two synthetic and three real-world datasets. The obtained results suggest that the proposed SOM-based classifiers consistently outperform standard supervised classifiers. © 2015 The Natural Computing Applications Forum

Abstract
Most of current physical-layer security techniques rely on a degraded eavesdropper, thus warranting some sort of advantage that can be relied upon to achieve higher levels of security. We consider instead non-degraded eavesdroppers, that possess equal or better capabilities than legitimate receivers. Under this challenging setup, most of current physical-layer security techniques become hard to administer and new dimensions to establish advantageous periods of communication are needed. For that, we characterize the secrecy level of two schemes for physical-layer security under non-degraded eavesdroppers: a spread spectrum uncoordinated frequency hopping scheme, and a jamming receiver with self-interference cancellation.

Abstract
An extension of the well-known Particle Swarm Optimization (PSO) to multi-robot applications has been recently proposed and denoted as Robotic Darwinian PSO (RDPSO), benefited from the dynamical partitioning of the whole population of robots. Although such strategy allows decreasing the amount of required information exchange among robots, a further analysis on the communication complexity of the RDPSO needs to be carried out so as to evaluate the scalability of the algorithm. Moreover, a further study on the most adequate multi-hop routing protocol should be conducted. Therefore, this paper starts by analyzing the architecture and characteristics of the RDPSO communication system, thus describing the dynamics of the communication data packet structure shared between teammates. Such procedure will be the first step to achieving a more scalable implementation of RDPSO by optimizing the communication procedure between robots. Second, an ad hoc on-demand distance vector reactive routing protocol is extended based on the RDPSO concepts, so as to reduce the communication overhead within swarms of robots. Experimental results with teams of 15 real robots and 60 simulated robots show that the proposed methodology significantly reduces the communication overhead, thus improving the scalability and applicability of the RDPSO algorithm.