Projects

(Preliminary research focus) Applications of 5G communication in vertical industries will require Ultra Reliable Low Latency Communication (URLLC) that was unprecedented in the previous generation mobile systems. On the other hand, the overall security mechanisms in 5G system, such as confidentiality protection, integrity protection, replay protection, and privacy protection are largely unchanged from the previous generation systems. It is ...

Over the past few years, the number as well as the sophistication of security threats has been increasing. State-of-the-art security analytics solutions are therefore increasingly depending on machine learning methods to distinguish malicious from normal behavior in diverse contexts. However, with the use of machine learning for security applications on the rise, so are the attacks of cyber- criminals on these methods. Furthermore, there is ...

It is estimated that 20-40% of expenses in healthcare have to do with wastefulness. For fraud this percentage can go up to 10%.
The goal of the PhD is to apply advanced data mining and machine learning techniques to discover new insights from the databases of health care funds.

Solar flares are huge bursts of electromagnetic radiation, and are the result of the complex and dynamic magnetic field of the Sun. If the intensity of light is monitored over time, then flares can be seen as a rapid rise of the intensity, followed by a more gradual decay. Solar flares are commonly observed to produce a series of regularly occurring pulses of light, known as quasiperiodic pulsations (QPPs), and these QPPs have been seen ...

One of the biggest challenges in current space weather research is to predict the characteristics that the high speed solar wind streams will have when arriving at the Earth (or at other planets) based on properties of their source, the coronal holes observed at the Sun. The innovative simulation tool EUHFORIA (EUropean Heliospheric FORecasting Information Asset, Pomoell and Poedts, 2018) is a new heliospheric 3D MHD model that has been ...

'With great power comes great responsibility' is a cliché that perfectly applies to artificial intelligence (AI). As technology is progressing faster than ever before towards software with more and more abilities, adaptation and autonomy, the risks are becoming increasingly apparent. As a matter of fact, humans are still lacking the tools to systematically obtain strong guarantees - about their safety, privacy, etc. - from the intelligent ...

This project aims at developing new methodologies for statistical inference for flexible mean and quantile regression models based on complex data, in particular censored data. The emphasis in this research will be on inference for quantiles, including quantile regression. In recent years, a few large asymmetric families of distributions have been studied. The interesting feature of these families is that the location parameter coincides with ...

The focus of the research is representing uncertainty in visual analytics applications as well as increasing user trust in predictions. The overall objective is to research meaningful visualisations for large and heterogeneous datasets, taking into account the aspects of data quality and uncertainty. The propagation of uncertainty has been researched by the visual analytics community under the umbrella of “data wrangling”. Most of this work, ...

AIDA brings a transformational innovation to the analysis of heliophysics data in four steps.
First, AIDA will develop a new open source software called AIDApp written in Python (a free language) and capable of
collecting, combining and correlating data from different space missions. AIDApp wants to replace mission-specific tools
written for costly languages (such as IDL) that exclude many scientists, students and amateur space ...

This research is situated in the field of automated dynamical decision making under uncertainty, a field with many applications, ranging from building financial portfolios, over stabilizing chemical plants, to controlling self-driving cars. In control, it is standard practice to use dynamical models to predict expected future behavior of the controlled system and compute control actions to achieve a specified goal. The ever-increasing need ...