Analysis of events data (PhD/Postdoc)
Event modelling concerns the analysis of ordered, timestamped data, with discrete states (or patterns) occurring at a particular point or points in the sequence. Examples include detecting faults or cyber-attacks in industrial processes and computer networks, predicting risk of fractures or disease flares in patients, learning of socio-political events through analysis of text data and even modelling the rate of taxi demand and hospital admissions.
We invite enthusiastic individuals to join our group at TU Kaiserslautern as a PhD student or postdoctoral researcher to work on a topic in this research area. Candidates should be interested in undertaking a project in the sphere of ’events data’, which includes one or a combination of the following sub-domains:
- event detection
- pattern recognition, anomaly and changepoint detection and process control: determining the occurrence events through recognizing patterns in historical data
- event prediction
- forecasting events, joint modelling of time-to-event and longitudinal data
- causal event modelling
- using structural models to answer counterfactual questions about the occurrence/non-occurrence of events
More information to follow.
If you would like to know more about this opportunity—both at PhD and at postdoctoral level—please get in touch.
Sebastian Vollmer
Professor for Applications of Machine Learning
My research interests lie at the interface of applied probability, statistical inference and machine learning.