Research Team I2S

Intelligent Information Systems

Since 1999, the I2S research team performs application-oriented research and development of intelligent information systems in the Internet and Web. Main focus is on the application of Artificial Intelligence (AI), Agent Technologies, Services Computing, and Semantic Web Technologies for innovative means of intelligent data analysis and service coordination.

Vision. Agents and semantic technologies, each with deep roots in AI, together with services computing are commonly considered as key enablers for the development of intelligent information systems in the future Internet. The future Internet is expected to arise from the ongoing convergence of the Internet of Services (IoS), the Internet of Things (IoT), the Internet of Interaction, and the Quantum Internet. Intelligent information systems in the future Internet are envisioned to provide users with assisted real-time access to relevant data and service coordination anywhere, anytime, on any device, to anyone, safe and secure, and in an intelligent way. Agent technologies for individual or joint AI planning and learning, inter-agent communication and coordination can be used to implement autonomous, proactive, rational, adaptive and collaborative behaviour of such systems with intelligent agents. Semantic technologies enable the semantic interoperation and coordination of distributed heterogeneous data, services, and processes by these agents for intelligent information systems in support of industry and business, as well as our private and social life.

Research. Research of the I2S team on intelligent service coordination addresses the challenges of scalable, flexible, situated optimal and high-performance semantic service discovery, selection, and composition planning by intelligent agents in dynamic and distributed environments. In case of contested or chargeable data and services, the agents may also have to enter rational and trusted negotiations (bargaining, coalition formation, auctions) in order to resolve the conflicting goals with maximized individual or social welfare of the respective multi-agent system. Research of the I2S team on intelligent data analysis addresses the challenges of scalable, privacy-aware distributed data mining, qualitative data (stream) analysis and semantic explanations of results to the user, approximated NP-hard AI and optimisation problem solving on GPU and in the cloud, and semantic deep learning for scene interpretation and planning by autonomous systems.

Applications. The research results of the I2S team are implemented in a wide range of free open-source and partly awarded tools and benchmarks for semantic service coordination and data analysis, as well as in prototypes of intelligent information systems for various application sectors. These sectors include

  • medical emergency assistance services in health care
  • smart event management and social media
  • decentralized power regulation in smart micro-grids
  • agricultural production planning and logistics
  • intelligent simulation of virtual production lines
  • process optimisation in cloud manufacturing
  • machine condition monitoring and maintenance
  • adaptive business processes in smart retail environments 
  • simulation of adaptive autonomous driving in virtual environments

Selected system prototypes were demonstrated at the CEBIT fair, the Hannover Industry fair, and the international Consumer Electronics Show (CES).

Since 2008, the research team is part of the DFKI Agents and Simulated Reality department (formerly: DFKI Deduction and Multiagent Systems department) and successfully contributed to many funded R&D projects in its core research areas.


PD. Dr. Matthias Klusch, email: klusch@dfki,de, phone: +49-681-85775-5297, fax: +49-681-85775-2235, postal address: German Research Center for Artificial Intelligence GmbH, Stuhlsatzenhausweg 3, D-66123 Saarbruecken, Germany