Current Projects
ADEWaS
In the ADEWaS project, which is financed by the Deutsche Telekom AG, a software system to automatically recognize anomalies in large amounts of data is to be developed in collaboration with the Deutsche Telekom Laboratories (T-Labs), the Ben-Gurion University of the Negev, the DAI lab of the TU Berlin and T-Systems South Africa. In the context of telecommunication services and infrastructure, critical conditions of infrastructure components (e.g. central servers) as well as strange behaviour of users can be considered as relevant anomalies. Due to the early recognition of these anomalies, the availability of the services can be guaranteed on the one hand, while on the other hand a possible misuse of the services is also recognized and prevented. Whereas up to now only simple rule-based approaches have been used, in ADEWaS an intelligent system will be developed which can recognize anomalies sooner and more reliably.
I am the leader of DFKI's workpackages within the industry project ADEWaS financed by Deutsche Telekom AG. The DFKI is developing software components for statistical anomaly detection within the framework of ADEWaS. In addition to statistical outlier detection methods, methods using non-supervised machine-learning are also applied. In order to be able to demonstrate the practical implementation of these algorithms, they will be evaluated using several real anomaly-recognition scenarios of the Deutsche Telekom AG.
INBEKI
The goal of the project INBEKI is the development of an integrated system for supporting law enforcement agencies in the area of child sexual abuse investigation. It is funded by the Federal Ministry of Eduction and Research (BMBF) over 3 years and is conducted by a consortium of five academic, industrial, and public partners.
I am the leader of DFKI's sub-project within the project INBEKI in which we are developing novel image analysis technologies for scene recognition, part image search, and text-in-image recognition.
iCOP
The iCOP project is targeted at fighting child pornography in P2P networks. A forensic toolkit is developed that supports police investigators of child sexual abuse with the identification of illegal image and video material and its originators. In the project, the DFKI develops multimedia analysis technology that automatically identifies child pornographic content. iCOP is funded through the EU Safer Internet Program.
myCBR
Together with Thomas
Roth-Berghofer I am developing the case-based reasoning tool myCBR.
Past Projects
FIVES
I was the project leader of DFKI's workpackages within the EU project FIVES
funded under the Safer Internet Plus program. The goal of FIVES was the
development of novel investigative tools specifically tailored for
investigations involving images and videos of child sexual abuse.
PaREn
The PaREn Project targeted on the development of novel tools allowing non-experts to use, train, test, and deploy pattern recognition and machine learning modules in real-world software systems.
NetShield
From 2005 to 2008 I was the leader of the industry project NetShield. The goal of the project was the development of a novel system for
detecting and mitigating Distributed Denial-of-Service (DDoS) attacks based on pattern recognition techniques. The project was part of the
NetCentric Security
project funded by the
Deutsche
Telekom Laboratories.
IPeT
The long-term goal of the project IPeT (Image-based Personal Computing Technologies), funded by
the german federal ministry of education and research, was to bridge
the gap between human and computer abilities to interpret image data
and to make image understanding technologies an integral part of personal computing.
The initial goal of this work was to create a flexible toolbox for document layout
analysis, optical character recognition, handwriting recognition, image matching, and
content-based retrieval based.
ZERBERUS
In the project ZERBERUS case-based reasoning technology was applied to avoid special bacterialogical problems
in sewage plants in cooperation with the Institute
of Environmental Engineering of the University
of Kaiserslautern and the Center
of Innovative WasteWater Technology tectraa. The project was
sponsored by the ministry of the environment of Rheinland-Pfalz.
SKe
In the project SKe ("Durchgängige Sicherheitskonzeption mit dynamischen
Kontrollmechanismen für eService Prozesse") I was working in close collaboration
with the Fraunhofer Institut
für Experimentelles Software Engineering (IESE). The project was funded by
the german federal ministry for education and research.
FIORES-II As a scientific assistant in the research group "Artificial Intelligence - Knowledge-Based Systems" I was involved in the EU-Project FIORES-II. Here, CBR-technology was applied in the area of aesthetic design.
WEBSELL
The goal of the EU ESPRIT project WEBSELL was the development of new intelligent sales support technology and related
products/tools for the World Wide Web. In the scope of my diploma thesis my contribution to WebSell was the development of a
novel product configuration system.
INRECA-II
As a student assistant I was involved in the EU ESPRIT-IV project INRECA-II ("Information and Knowledge Reengineering for Reasoning from Cases") where
I implemented graphical similarity editors and a rule interpreter for the commercial CBR-Tool
CBR-Works.