Intelligent Systems

There are many definitions of what an "intelligent system" is. Clearly, the characterization "intelligent" allows for various interpretations. To distinguish it from the research areas of learning systems and autonomous agents as well as multi-agent systems (both of which can be said to be intelligent as well), my interpretation of intelligent systems emphasizes their supportive and assistive character. These systems are meant to be used by humans, they are meant to help them in accomplishing some task, and they are meant to do so in a smart and intelligent way.

Along these lines, I have been involved in projects that revolve around the following topics.

  • decision support systems and recommender systems
  • case-based reasoning and experience management
  • knowledge-based systems and knowledge management
  • intelligent search and semantic web

Project Spotlights

Semantically Enabled Knowledge Technologies (SEKT)

The SEKT project developed and exploited semantic knowledge technologies. Core to this EU IST project has been the creation of synergies from combining approaches from the research areas of ontology management, machine learning and natural language processing. SEKT envisioned a knowledge workplaces where the boundaries between document management, content management, and knowledge management are broken down, and where knowledge management is an effortless part of day to day activities. Appropriate knowledge is automatically delivered to the right people at the right time at the right granularity via a range of user devices.
Visit the project website.


This project investigated AI-based techniques as tools for environmental engineering. One such area is the optimization of processes in wastewater treatment plants. The experience and knowledge that human managers of such plants have gathered from past events is of crucial importance in that domain. In Zerberus, an approach based on Case-Based Reasoning was investigated that helped identifying harmful microorganisms and a decision support system was implemtented that suggested possible counteracting measures.
Visit the project website.

Case-Based Reasoning

The basic assumption underlying the research field of case-based reasoning (CBR) is that similar problems do have similar solutions. When being faced with a new problem, the core idea of CBR is to turn to a collection of past experiences, so-called cases, search for one that appears to be similar to the current one and to re-use (possibly with slight adaptations) the solution part of the former case.

Case-based reasoning is both, a powerful method for reasoning with computers, as well as a description of human behavior in everyday problem solving. With respect to the former, CBR has been formalized as a four-step process consisting of the stages retrieve, revise, reuse, and retain (see figure).

My main interests in the realm of case-based reasoning center around the modeling and usage of similarity measures which are an essential part of the first phase. Using machine learning techniques for learning and optimizing such measures has been the topic of research described here. Besides, I am also interested in the use of CBR methods as a tool for other disciplines, their application to practical problems, and I have done some work on making formal statements on the precision of case-based inference (see the following publication).

Arrival Manager AMAN

The arrival manager AMAN is a decision support system for air traffic controllers. It is capable of registering all arriving aircraft within a range of up to 200 miles and more around the airport. It continuously calculates hand-over and landing times on the basis of different parameters, such as speed, current and predicted flight path, traffic density and air traffic controller activities. Based on these information the software calculates an optimal approach sequence that is displayed and recommended to the human user. This way the AMAN system supports air traffic controllers in meeting safety requirements while enabling the maximum utilisation of high density airspaces. AMAN is developed by the German Air Navigation Provider (Deutsche Flugsicherung).