Title : Ambient Agent & Behavior Change Intervention: The Outreach

Presenter Azizi Ab Aziz
Abstract Ambient Intelligence (AmI) refers to a new paradigm in computing, in which people are empowered through a digital environment that is aware of their presence and context, and is responsive to their needs. Current trends within AmI technology provides a wider spectrum of opportunities to be implemented in personal care applications. To serve better personal care in AmI, we need technology can that is aware of humans; an agent that uses a computational model to reason about humans can be used for this. Derived by this motivation; I will talk about my inspiration on how to have such agents that capable to “assist” human in behavioral change intervention (BCI) , with a special focus on affective disorder domain. Behavioral change intervention can be defined as a process of rapid and involuntary change of behavior associated with a mental disorder and addiction. Many models derived from BCI are attempted to explain the reasons behind alterations in individuals' behavioral patterns. These theories explain how environmental, personal, and behavioral characteristics as the major factors in behavioral determination. Combining this perspective with ambient agent will give a new spectrum of change interventions process. The focal idea of this presentation can be divided into three parts. For the first part of the talk, I will cover some issues in affective and behavioral disorders, such as important aspects in the progression of affective disorder, relapse, and recurrence. In the second part, special attention on how ambient agents can be used as a medium to provide a better behavioral change interventions program will be presented. During this part, some recent and previous works will be presented. Main concepts in behavioral change models and how it can be integrated with ambient agents are also within the discussion. Finally, for the last part of my presentation, I will partially discuss the on-going progress of my research work, and challenges that may lie ahead.

Title : Fraud Detection at the VU: three case studies

Presenter Wojtek Kowalczyk
Abstract During the last year I worked on 3 projects aimed at detecting fraud in various sorts of data: - health insurance - US Crop Insurance Program - electronic payments In my talk I will discuss these cases in some detail, focusing mainly on efficient data structures that make the detection process very efficient. In particular, I will discuss static hash tables and multi-linked rotating buffers that I used in a prototype system that monitors, in real time, 1 billion (1000.000.000) of the most recent transactions (using the sliding window model). Additionally, I will share some experience with using kd-trees for analyzing spatial data from the US Department of Agriculture.