Der Einsatz von Künstlicher Intelligenz zur Betrugserkennung bei Heilmittel- und Pflegeleistungen in der gesetzlichen Krankenversicherung. Von der Pseudonymisierung und Digitalisierung der Abrechnungsbögen bis zur Analyse, Auswertung und Darstellung der Anomalien - Ein Projektbericht! 

Zielpublikum: Management, Data Scientists, Data Engineers 
Voraussetzungen: Experience, Curiosity 
Schwierigkeitsgrad: Fortgeschritten

The Coronavirus lockdowns altered public transport occupation data. Ultimately, these changes in occupation data are perfect examples of sudden concept drifts that can be blockers in most machine learning deployments. We managed to overcome the obstacles by developing methods and engineering features that allowed us to adjust forecasts based on unforeseen changes in the occupation data. In this talk, we give insights into our journey from idea development to the ways how we overcame the challenges and share our learnings. 

Target Audience: Data Engineers, Software Architects, AI Architects, Data Scientists 
Prerequisites: Experience in Time Series. Basic understanding of machine learning 
Level: Expert 

Extended Abstract: 
Machine learning projects often view and predict a snapshot interval of reality. We machine learning engineers often forget that the real world is not static at all.  

After all, we got hit by the Coronavirus reality rendering countless machine learning models useless.  

The changes in public transport occupation out of Corona lockdowns is a perfect example of these so-called concept drifts. 

The instability of the machine learning models when concept drifts are appearing result in immense problems for the reliability and predictability of the AI.  

Therefore, we dig into this problem and show how we solved this challenge for the case of concept dirfts in public transport occupations.