PhD Defence Mandani Ntekouli

Keywords: Ecological Momentary Assessment, time-series clustering, group-based modeling 
 

"Bridging Individual and Group Perspectives in Psychopathology: Computational Modeling Approaches using Ecological Momentary Assessment Data"

Mado Ntekouli’s thesis bridges the knowledge of psychology and data science to develop advanced analysis methods aimed at improving the understanding of mental disorders. Mental disorders are inherently complex because everyone experiences emotions and behavior in their own way. Therefore, traditional research models often struggle to capture this complexity, relying on simplistic relationships between variables. 

This research leverages machine learning and data from Ecological Momentary Assessment, a method that measures emotions, behaviors and experiences in real-time rather than retrospectively. Initially, the focus is on developing personalized models tailored to individual’s unique patterns. Furthermore, the research intelligently combines data from multiple participants to enhance the model’s predictive power. Through clustering, groups with similar characteristics can be identified and improve model performance. 

What does this mean in practice? This work could lead to better assessment and treatment of mental disorders and contribute to a deeper understanding of mental health conditions. 

Click here for the live stream.