Search

Reading is one of the essential processes during the maturation of an individual. It is estimated that 5-10% of school-age children are affected by dyslexia, the reading disorder characterised by difficulties in the accuracy or fluency of word recognition. There are many studies which have reported that coloured overlays and background could improve the reading process, especially in children with reading disorders. As dyslexia has neurobiological origins, the aim of the present research was to understand the relationship between physiological parameters and colour modifications in the text and background during reading in children with and without dyslexia. We have measured differences in electroencephalography (EEG), heart rate variability (HRV), electrodermal activities (EDA) and eye movements of the 36 school-age (from 8 to 12 years old) children (18 with dyslexia and 18 of control group) during the reading task in 13 combinations of background and overlay colours. Our findings showed that the dyslexic children have longer reading duration, fixation count, fixation duration average, fixation duration total, and longer saccade count, saccade duration total, and saccade duration average while reading on white and coloured background/overlay. It was found that the turquoise background, turquoise overlay, and yellow background colours are beneficial for dyslexic readers, as they achieved the shortest time duration of the reading tasks when these colours were used. Additionally, dyslexic children have higher values of beta (15-40 Hz) and the broadband EEG (0.5-40 Hz) power while reading in one particular colour (purple), as well as increasing theta range power while reading with the purple overlay. We have observed no significant differences between HRV parameters on white colour, except for single colours (purple, turquoise overlay, and yellow overlay) where the control group showed higher values for mean HR, while dyslexic children scored higher with mean RR. Regar

In this paper, we consider the symmetric multi-type non-negative matrix tri-factorization problem (SNMTF), which attempts to factorize several symmetric non-negative matrices simultaneously. This can be considered as a generalization of the classical non-negative matrix tri-factorization problem and includes a non-convex objective function which is a multivariate sixth degree polynomial and a has convex feasibility set. It has a special importance in data science, since it serves as a mathematical model for the fusion of different data sources in data clustering. We develop four methods to solve the SNMTF. They are based on four theoretical approaches known from the literature: the fixed point method (FPM), the block-coordinate descent with projected gradient (BCD), the gradient method with exact line search (GM-ELS) and the adaptive moment estimation method (ADAM). For each of these methods we offer a software implementation: for the former two methods we use Matlab and for the latter Python with the TensorFlow library. We test these methods on three data-sets: the synthetic data-set we generated, while the others represent real-life similarities between different objects. Extensive numerical results show that with sufficient computing time all four methods perform satisfactorily and ADAM most often yields the best mean square error (MSE). However, if the computation time is limited, FPM gives the best MSE because it shows the fastest convergence at the beginning. All data-sets and codes are publicly available on our GitLab profile., The authors acknowledge the financial support from the Slovenian Research Agency (research core funding No. P2-0098, and projects No. J1-8155, No. PR-07606,No. N1-0071), from the European Research Council (ERC) Consolidator Grant (grant number 770827) and from the Spanish State Research Agency AEI 10.13039/501100011033 (grant number PID2019-105500GB-I00)., Peer Reviewed, Postprint (author's final draft)

The Big Data and data science revolution, internet of things, advances in machine learning, improvements in wireless connectivity, sensor technologies and available computing power. This final chapter takes a step back and presents some views towards the future. Whether one deals with PM on a manufacturing process or a fleet of machines, in logistics or construction, the potential gains from improving the maintenance policies can be substantial. The chapter focuses on the following provocative questions: Is it cybernetic or is it human? Real-time communication in maintenance? How to determine granularity in space and time? Open or closed maintainability? Insourcing or outsourcing? Explicit modeling or data-driven pragmatics? How to apply Virtual Reality and Augmented Reality? Service robotics for maintenance? Today manufacturing companies are very keen on developing maintenance service businesses for the equipment they have sold.

Data analysis is of paramount importance in the management of proactive maintenance. This chapter provides a deep analysis of the different techniques that can be adopted when dealing with the automation of maintenance processes. In particular, it focuses on three aspects of the maintenance that we have considered as the cornerstones of PM: root cause analysis, identification of the remaining useful life, and alerting and predicting of failures. Data can be extracted and stored from machines and components. Machine based data analysis may be used locally to optimize the machine’s maintenance. Better results may be expected if these particular data are combined with other data sources such as that from the MES, ERP, cloud-based open-data sources and the machine operators. One of the subsystems in a sheet metal machine is the hydraulics subsystem. Domain expertise is used to define a set of hydraulics failure modes, which are easy to understand and easy to describe.

The aim of the MANTIS project is to provide a proactive maintenance service platform architecture based on Cyber Physical Systems. The platform will allow estimating future performance, predicting and preventing imminent failures and scheduling proactive maintenance. Maintenance is an important element that creates added value in the business processes and it also creates new business models with a stronger service orientation. Physical systems and the environment they work in are continuously monitored by a range of intelligent sensors, resulting in massive amounts of data, which characterise the usage history, working condition, location, movement and other physical properties of the systems. These systems are part of a larger network of heterogeneous and collaborative systems (e.g. vehicle fleets) connected via robust communication mechanisms able to operate in challenging environments. MANTIS consists of distributed processing chains that efficiently transform raw data into knowledge, while minimising the need for bandwidth. Sophisticated distributed sensing and decision-making functions are performed at different levels collaboratively, ranging from local nodes to locally optimise performance, bandwidth and maintenance; to cloud-based platforms that integrate information from diverse systems and execute distributed processing and analytics algorithms for global decision-making.