A significant proportion of patients with drug-resistant epilepsy that are treated with resective surgery does not achieve seizure freedom. In most cases, surgical failure is due to inaccurate localization of the epileptogenic zone (EZ). The goal of this project is to develop new techniques for the analysis of scalp and intracranial EEG that will allow for a more reliable localisation of the EZ, which will in turn result in improved outcome after resective epilepsy surgery. We will do so by developing techniques for EEG analysis based on machine learning, artificial intelligence and concepts from information theory (interaction information, partial information decomposition). We will also extend our previous work on effective connectivity analysis on ictal intracranial EEG. This work will have an impact on clinical practice and improve our understanding of network dynamics in epilepsy.