Abstract
The majority of evolving data streams classification algorithms assume that the actual labels of the predicted examples are readily available without any time delay just after a prediction is made. However, given the high label costs, dependence of an expert, limitations in data transmission or even restrictions imposed by the problem's nature, there is a large number of real-world applications in which the availability of actual labels is infinitely delayed (never available), In these cases, it is necessary the use of algorithms that does not follow the traditional process of monitoring the error rate to detect changes in data distribution and uses the most recent labeled data to update the classification model. In this paper, we propose the method Maasstfication to classify evolving data streams with infinitely delayed labels. Our method is inspired on the use of Micro-Cluster representation from online clustering algorithms. Considering the presence of incremental drifts, our approach uses a distance-based strategy to maintain the Micro-Clusters' positions updated. An evaluation in several synthetic and real data shows that Maassification achieves competitive accuracy results to state-of-the-art methods and adequate computational cost. The main advantage of the proposed method is the absence of critical parameters that require user's prior knowledge, as occurs with rival methods.

Abstract
One of the most important disadvantages of photocatalytic treatments is the high cost associated with the use of UV lamps. In this work, the efficiency of two UV-A LEDs (ultraviolet-a light emitting diodes) photosystems as a low cost alternative to conventional UV lamps was tested. The efficiency of the two UV-A LEDs photosystems was compared to that of the most economical UV source-solar radiation. To this end, the oxidative discolouration of Reactive Black 5 (RB5) aqueous solutions was studied using photocatalysis with different concentrations of TiO2 (0.5-1 g/L) and H2O2 (1.76, 4.41, 8.82 and 17.64 mM), exposed to different radiation sources: UV-A LEDs and solar radiation. The use of H2O2 increased the discolouration rate of RB5; however, an excessive dosage reduced the yield of the treatment, and the best results were attained with a concentration of 8.82 mM of H2O2. Strong differences were observed between the use of UV-A LEDs (23 W/m(2)) and solar radiation. In both cases total discolouration was observed, but the discolouration rate was considerably higher with solar radiation. However, the use of a more powerful UV-A LED photo-system (85 W/m(2)) allowed the achievement of higher discolouration rates (k = 0.284 min 1) than those obtained with solar radiation (k = 0.189 min 1) using only 0.5 g/L of TiO2. Therefore, UV-A LED radiation is a serious alternative to conventional UV lamps, since they are ecofriendly, have a low operational cost and high energy efficiency.

Abstract
In this paper, we describe how a learning platform that takes into account the student's learning preferences, the personality and emotions could prompt better learning results. This model will assess the emotional state of the student in an online learning environment by introducing techniques of Affective Computing that can capture the student emotional state and based on that, adapt the course to the characteristics and needs of the student in order to get an improvement in the learning results. Students, as individuals, differ in their social, intellectual, physical, psychological, emotional, and ethnic characteristics. Also, differ in their learning rates, objectives and motivation turning, their behaviour rather unpredictable. Added to emotion, and in order to obtain an effective model for online learning, student's personality and learning style are also considered. The architecture developed was tested by a group of students of higher education in Oporto. The results indicated that the model created used can support and improve the student's results, verifying that a negative emotional state could influence the learning process. © 2015 ACM.

Abstract
Dance movement is intrinsically connected to the rhythm of music and is a fundamental form of nonverbal communication present in daily human interactions. In order to enable robots to interact with humans in natural real-world environments through dance, these robots must be able to listen to music while robustly tracking the beat of continuous musical stimuli and simultaneously responding to human speech. In this paper, we propose the integration of a real-time beat tracking system with state recovery with different preprocessing solutions used in robot audition for its application to interactive dancing robots. The proposed system is assessed under different real-world acoustic conditions of increasing complexity, which consider multiple audio sources of different kinds, multiple noise sources of different natures, continuous musical and speech stimuli, and the effects of beat-synchronous ego-motion noise and of jittering in ego noise (EN). The overall results suggest improved beat tracking accuracy with lower reaction times to music transitions, while still enhancing automatic speech recognition (ASR) run in parallel in the most challenging conditions. These results corroborate the application of the proposed system for interactive dancing robots.

Abstract
In recent years the focus of research in the fields of iris and face recognition has turned towards alternative traits to aid in the recognition process under less constrained acquisition scenarios. The present work assesses the potential of the periocular region as an alternative to both iris and face in such conditions. An automatic modeling of SIFT descriptors, using a GMM-based Universal Background Model method, is proposed. This framework is based on the Universal Background Model strategy, first proposed for speaker verification, extrapolated into an image-based application. Such approach allows a tight coupling between individual models and a robust likelihood-ratio decision step. The algorithm was tested on the UBIRIS. v2 and the MobBIO databases and presented state-of-the-art performance for a variety of experimental setups.

Abstract
In a time of economic austerity, more pressure is being put on the existing transport systems to be more sustainable and, at the same time, more equitable and socially inclusive. Regular public road transportation traditionally uses fixed routes and schedules, which can be extremely expensive in rural areas and certain periods of the day in urban areas due to low and unpredictable demand. Demand Responsive Transportation systems are a kind of hybrid transportation approach between a taxi and a bus that try to address these problems with routes and frequencies that may vary according to the actual observed demand. Demand Responsive Transportation seems to have potential to answer the sustainability and social inclusion challenges in a context of austerity. However, DRT projects may fail: it is not only important to solve the underlying model in an efficient way, but also to understand how different ways of operating the service affect customers and operators. To help design DRT services, we developed an innovative approach integrating simulation and optimization. Using this simulator, we compared a real night-time bus service in the city of Porto, Portugal, with a hypothetical flexible DRT service for the same scenario.