Ricerc@Sapienza

Interpretability, accuracy and a solid neurophysiological basis can be considered as the main requirements for the classification model to monitor motor imagery tasks in post-stroke motor recovery paradigms supported by the brain-computer interface technology. This study aimed at comparing the accuracy performance of different classification approaches applied on a dataset of 15 stroke patients. We also explored how the variation in the dimensionality of the feature domain would influence the different classifier performance.

Educational Institutions data constitute the basis for several important analyses on the educational systems; however they often contain not negligible shares of missing values, for several reasons. We consider in this work the relevant case of the European Tertiary Education Register (ETER), describing the Educational Institutions of Europe. The presence of missing values prevents the full exploitation of this database, since several types of analyses that could be performed are currently impracticable.

The imputation of missing values in the detail data of Educational Institutions is a difficult task. These data contain multivariate time series, which cannot be satisfactory imputed by many existing imputation techniques. Moreover, almost all the data of an Institution are interconnected: the number of graduates is not independent from the number of students, the expenditure is not independent from the staff, etc. In other words, each imputed value has an impact on the whole set of data of the institution.

We propose here an implementation of 2D hierarchical clustering tailored for power constrained and low-precision hardware. In many application fields such as smart sensor networks, having low computational capacity is mandatory for energy saving purposes. In this context, we aim to deploy a specific constrained hardware solution, using a parallel architecture with a low number of bits. The effectiveness of the proposed approach is corroborated by testing it on well-known 2D clustering datasets.

In this paper we propose a novel hardware implementation for a bidimensional unconstrained hierarchical clustering method, based on fuzzy logic and membership functions. Unlike classical clustering approaches, our work is based on an advanced algorithm that shows an intrinsic parallelism. Such parallelism can be exploited to design an efficient hardware implementation suitable for low-resources, low-power and highcomputational demanding applications like smart-sensors and IoT devices. We validated our design by an extensive simulation campaign on well known 2D clustering datasets.

Modeling and predicting failures in the field of predictive maintenance is a challenging task. An important issue of an intelligent predictive maintenance system, exploited also for Condition Based Maintenance applications, is the failure probability estimation that can be found uncalibrated for most standard and custom classifiers grounded on Machine learning.

In this paper we investigate a public Fast Charge (FC) station nanogrid equipped with a Photovoltaic (PV) system and an Energy Storage System (ESS) using second-life Electric Vehicle (EV) batteries. Since the nanogrid is intended for installation in urban areas, it is designed with a very limited connection with the grid to assure peak shaving and encourage PV autoconsumption.

The high time needed to reconfigure cloud resources in Network Function Virtualization network environments has led to the proposal of solutions in which a prediction based-resource allocation is performed. All of them are based on traffic or needed resource prediction with the minimization of symmetric loss functions like Mean Squared Error. When inevitable prediction errors are made, the prediction methodologies are not able to differently weigh positive and negative prediction errors that could impact the total network cost.

Machine Learning (ML) algorithms have become popular in many fields, including applications related to turbomachinery and heat transfer. The key properties of ML are the capability to partially tackle the problem of slowing down of Moore’s law and to dig-out correlations within large datasets like those available on turbomachinery. Data come from experiments and simulations with different degree of accuracy, according to the test-rig or the CFD approach.

Ventilation systems are used in gas turbine packages to control the air temperature, to protect electrical instrumentation and auxiliary items installed inside the enclosure and to ensure a proper dilution of potentially dangerous gas leakages. These objectives are reached only if the ventilation flow is uniformly distributed in the whole volume of the package, providing a good air flow quality as prescribed by international codes such as ISO 21789.