HANNOVER MESSE 2020,
20 - 24 April
Digitalization leads to new opportunities for setting prices dynamically. Dynamic Pricing problems are fundamentally different and cannot be handeled with methods of prediction analytics. Strictly speaking, it is a "causal" problem. We develop a structural model for consumer demand and estimate it with Machine Learning methods. This allows us to estimate the effect of different pricing schemes on demand. Consistent estimation of the demand function is core for profit or revenue maximization. We demonstrate the superior performance of our system in a real world problem of dynamic pricing of ride sharing services.
Prediction of orders is core for an efficient management of the supply chain and for planning. Machine Learning methods are superior for such prediction tasks. We ...
Fraud Detection is important for many companies. We develop a data-driven fraud detection system based on Deep Learning.