Abstract
The present paper describes the state of the art related to IIoT Devices and Cyber-Physical systems and presents a use case related to predictive maintenance. Industry 4.0 is the boost for smart manufacturing and demands flexibility and adaptability of all devices/machines in the shop floor. The machines must become smart and interact with other machines inside and outside the industries/factories. The predictive maintenance is a key topic in this industrial revolution. The reason is based on the idea that smart machines must be capable to automatically identify and predict possible faults and actuate before they occur. Vibrations can be problematic in electrical motors. For this reason, we address an experimental study associated with an automatic classification procedure, that runs in the smart devices to detect anomalies. The results corroborate the applicability and usefulness of this machine learning algorithm to predict vibration faults. © 2019, Springer International Publishing AG, part of Springer Nature.

Abstract
The continuous use of dynamic and disruptive ICT as energizing elements of the educational process is a reality of current days, where millennials are the centre of an education paradigm in which students are much more inclined to use technologies than enrolling in a traditional nondigital course. Considering education management information systems (EMIS) capacities to collect, analyse, process and publish information and data, it is easy to perceive their relevance to both education organizations and students. Nevertheless, and despite EMIS complexity and inherent possibilities, the existing literature does not provide for a detailed characterization on the impact these systems might have on students' success. Thus, this research focuses on understanding the use of EMIS by students and the arising of net benefits; it introduces an EMIS success model which posits that to ensure net benefits for students, education institutions must safeguard that their education management information systems are of high quality, while at the same time students are maintained satisfied with the system and engage in continuous use. To assess the posed model, an empirical study has been performed, involving students from higher education institutions. Findings from the study allow us to perceive that, as information systems (IS) success models state, EMIS use and students' satisfaction are predictors of net benefits. This same model also claims that the available information quality and EMIS inherent service quality are also strong determinants of both continuous EMIS use and student satisfaction.

Abstract
Background & aims: Bariatric surgery is increasingly common in the fight against morbid obesity. However, after this intervention, it is not fully understood the evolution of weight loss and how body composition changes. The objective of this work is to study the evolution after surgery of weight and body composition of obese patients that underwent bariatric surgery. Methods: In this retrospective and prospective study, we studied initially BMI and body composition of obese patients who attended nutritional appointments at Centro Hospitalar Sao Joao E.P.E. We collected personal data and anthropometric measurements between the pre-surgery appointment up to 60 months after surgery. Results: The sample consisted of 793 patients, of which 86.5% were female and 13.5% were male, with a mean age of 43 years (SD = 10.5 years) and mean height of 1.62 m (SD = 0.079 m). Patients undergoing gastric band, sleeve gastrectomy and gastric bypass had, respectively, an initial BMI reduction of 6.3 kg/ m(2), 13.2 kg/m(2) and 15.4 kg/m(2) and an initial fat mass% reduction of 4.4%, 14.3% and 17.3%. On the other hand, they had an initial increase of 3.2%, 10.8% and 12.4% of water%, 1.4%, 3.9% and 4.6% of fat and waterfree mass%, and 1.9%, 7.3% and 8.9% of skeletal muscle mass%, respectively. BMI and fat mass% on average had a large decrease in the first 12 months, increasing slightly from 24 months onwards. The opposite behaviour was observed for water%, fat and water-free mass% and skeletal muscle mass%. Conclusions: Bariatric surgery initially allows a substantial decrease in BMI as well as beneficial changes in the overall body composition of the individuals. Gastric bypass was the method that caused the most changes, followed by sleeve gastrectomy and, finally, gastric band. On average, after 24 months of follow-up, and for all surgical procedures studied, we observed a reversion in BMI and body composition values, showing the difficulties in maintaining weight and fat loss.

Abstract
Climate change is projected to be a key influence on crop yields across the globe. Regarding viticulture, primary climate vectors with a significant impact include temperature, moisture stress, and radiation. Within this context, it is of foremost importance to monitor soils' moisture levels, as well as to detect pests, diseases, and possible problems with irrigation equipment. Regular monitoring activities will enable timely measures that may trigger field interventions that are used to preserve grapevines' phytosanitary state, saving both time and money, while assuring a more sustainable activity. This study employs unmanned aerial vehicles (UAVs) to acquire aerial imagery, using RGB, multispectral and thermal infrared sensors in a vineyard located in the Portuguese Douro wine region. Data acquired enabled the multi-temporal characterization of the vineyard development throughout a season through the computation of the normalized difference vegetation index, crop surface models, and the crop water stress index. Moreover, vigour maps were computed in three classes (high, medium, and low) with different approaches: (1) considering the whole vineyard, including inter-row vegetation and bare soil; (2) considering only automatically detected grapevine vegetation; and (3) also considering grapevine vegetation by only applying a normalization process before creating the vigour maps. Results showed that vigour maps considering only grapevine vegetation provided an accurate representation of the vineyard variability. Furthermore, significant spatial associations can be gathered through (i) a multi-temporal analysis of vigour maps, and (ii) by comparing vigour maps with both height and water stress estimation. This type of analysis can assist, in a significant way, the decision-making processes in viticulture.

Abstract
To ensure high availability in large scale distributed systems, Conflict-free Replicated Data Types (CRDTs) relax consistency by allowing immediate query and update operations at the local replica, with no need for remote synchronization. State-based CRDTs synchronize replicas by periodically sending their full state to other replicas, which can become extremely costly as the CRDT state grows. Delta-based CRDTs address this problem by producing small incremental states (deltas) to be used in synchronization instead of the full state. However, current synchronization algorithms for delta-based CRDTs induce redundant wasteful delta propagation, performing worse than expected, and surprisingly, no better than state-based. In this paper we: 1) identify two sources of inefficiency in current synchronization algorithms for delta-based CRDTs; 2) bring the concept of join decomposition to state-based CRDTs; 3) exploit join decompositions to obtain optimal deltas and 4) improve the efficiency of synchronization algorithms; and finally, 5) experimentally evaluate the improved algorithms.

Abstract
Extreme and rare events, such as abnormal spikes in air pollution or weather conditions can have serious repercussions. Many of these sorts of events develop from spatio-temporal processes, and accurate predictions are a most valuable tool in addressing their impact, in a timely manner. In this paper, we propose a new set of resampling strategies for imbalanced spatiotemporal forecasting tasks, by introducing bias into formerly random processes. This spatio-temporal bias includes a hyperparameter that regulates the relative importance of the temporal and spatial dimensions in the selection of observations during under- or over-sampling. We test and compare our proposals against standard versions of the strategies on 10 different georeferenced numeric time series, using 3 distinct off-the-shelf learning algorithms. Experimental results show that our proposal provides an advantage over random resampling strategies in imbalanced spatio-temporal forecasting tasks. Additionally, we also find that valuing an observation's recency is more useful when over-sampling; while valuing its spatial distance to other cases with extreme values is more beneficial when under-sampling.