Time-series data is omnipresent in almost every domain. Think of physiological signals, and disease trajectories in Healthcare, IoT and monitoring data in Industry 5.0, energy consumption patterns in Energy, network traffic data in Telecommunications, etc. Across all these different application domains IDLAB focuses on:
Specific research track differentiators include explainability, i.e., enabling interpretable machine learning models, uncertainty quantification, taking into account both the variability in the outcome of an experiment which is due to inherently random effects (aleatoric) as well as the uncertainty caused by a lack of knowledge (epistemic) and sensor fusion.
IDLAB Ghent has expertise in (applied) research on machine learning for time-series data, mainly (but not exclusively) within the domains of:
Healthcare
e.g. prediction of disease progression (MS, respiratory & heart-related)
Industry 5.0
e.g. predictive maintenance, anomalies, prognostics & health monitoring
Cybersecurity
e.g. design of real-time network intrusion detection systems
Networks & Telecommunications
e.g. zero-touch management of wireless-wired networks
Energy
e.g. energy (peak) consumption prediction
Sports
e.g. automatic recognition of badminton strokes, positions & strategies
