Title : When love and hate collide: Dynamics in social support provision and receipt during stress/depression

Presenter Azizi Ab Aziz
Abstract This talk will give an overview of a social support provision and receipt model during the present of stress. The underlying principle of this model is a stress-buffering mechanism. Basically, the stress-buffering mechanism is a theoretical concept that proposes that social support protects or buffers individuals from the harmful effects of stress on health and well-being. It means that beneficial effects of social support can only occur when individuals are exposed to stress. An important mechanism behind this process is the selection of social support networks based on the strength of the tie with the people in the network. Social support receipt and provision components answer the question of individual’s preferences towards certain individuals within the social support networks. I will discuss those components and work that has been carried out using this stress buffering concept.

Title : Fraud detection

Presenter Wojtek Kowalczyk
Abstract During my presentation I will talk about our activities around detecting anomalies and fraud in data streams. In particular, I will talk about: - detecting skimming fraud (FDEC/Equens) - finding unexplainable outliers in insurance data (with Rob Konijn) - piecewise linear approximations of cpdf's (with Zoltan Szlavik).