Abstract
Predicting whether a student will pass or fail is one of the most important actions to take while giving lectures. Usually, the experienced teacher is able to detect problematic situations at early stages. However, this is only true for classes up to a hundred students. For bigger ones, automatic methods are needed. In this paper, we present a predictive system based on three criteria retrieved and computed from the logs of the learning management system. We built fast frugal decision trees to help predict and prevent student failures, using data retrieved from their resource usage patterns. Evaluation of the decision system shows that the system's accuracy is very high both in train and test phases, surpassing logistic regression and CART. © 2017 IEEE.

Abstract
A manufacturing system often consists of multiple units as workcells with complex work systems to achieve the desired outcomes in an efficient and effective manner. Uncertain events such as machine down time or scheduled maintenance are unavoidable in any manufacturing unit. In this paper, we are trying to find the maximum workload of the remaining machines to fulfill the production requirements. To achieve this, a dynamic workload adjustment strategy has been proposed with dynamic upgradation of residual life distribution model. With parallel configurations and different benchmark instances the simulation experiments has been conducted to evaluate the degradation rate of different units. Results show that the proposed method is effective for finding the residual life of multi-unit systems.

Abstract
Under specific conditions, there is a formal analogy between the fundamental equations of electromagnetism and relativistic gravitation, described by the Einstein field equations of general relativity. In this paper, we report on how we have used this analogy to implement a solver of the Einstein equations adapting algorithms initially developed for electromagnetism, combined with methods of heterogeneous supercomputing, in GPU that can achieve fast computing and exhibit good performance. We also present the results of the simulations used to validate our solver. © 2017 SPIE.

Abstract
Electricity markets worldwide are complex and dynamic environments with very particular characteristics. The markets' restructuring and evolution into regional and continental scales, along with the constant changes brought by the increasing necessity for an adequate integration of renewable energy sources are the main drivers. Multi-agent based software is particularly well fitted to analyse dynamic and adaptive systems with complex interactions among its constituents, such as electricity markets. This paper proposes the use of ontologies to enable the exchange of information and knowledge, to test different market models and to allow market players from different systems to interact in common market environments. Focusing, namely, on the EPEX electricity market.

Abstract
Early diagnosis is crucial in Diabetic Retinopathy (DR), to avoid further complications. The disease can be classified into one of two stages (an early stage of non-proliferative and a later stage of proliferative diabetic retinopathy), diagnosed based on existence and quantity of a characteristic set of lesions, such as micro-aneurysms, hemorrhages or exudates, in Eye Fundus Images (EFI). It is therefore important to segment adequately regions of potential lesions, to highlight and classify the lesions and the degree of DR. Density clustering methods are promising candidates to isolate individual lesions, and should be used together with effective techniques for vascular tree removal, feature extraction and classification. In this work we report on our approach, results, tradeoffs and conclusions for segmenting and detecting individual lesions. © 2017 IEEE.

Abstract
In this paper, we further explore the concept of phase-conjugated twin waves (PCTW) for nonlinear cancellation in space-division multiplexed (SDM) systems. Previously, we demonstrated that the PCTW technique can successfully provide nonlinear cancellation in SDM systems. In this paper, we investigate the cases where two and four spatial modes are copropagating in a multimode fiber, considering three link lengths (1000, 3200, and 8000 km). Weak-and strong-coupling regimes are also evaluated. Our numerical simulation results show an average performance improvement > 10 dB after a 1000 km transmission link.