Selecting the Right Individual Actions via Uplift Modeling in Dataiku DSS

19.03. | 10:15-10:45 | festival-stage

In Telecommunications as in most of the other industries, it is important to keep customers happy. In order to achieve this, it is very important to contact them with the right individual actions (offer, information, etc.) as well as not to contact them with the wrong ones.  Uplift modeling is a technique that can be used to determine which customer should be contacted with which action by predicting the individual causal effect. Based on the open “Criteo Uplift Modeling Dataset” we demonstrate how uplift modeling can be performed with the help of Dataiku DSS.

Speaker:

Dr. Björn Höfer (Telefónica Germany GmbH & Co. OHG)
Manuel Nitzsche (Dataiku SAS)