CLASS (Collaboration in LAnguage and Speech Science and technology) is an Accompanying Measure of the European Commission in FP5. It coordinates cluster and cross-cluster research among the RTD projects which have been established following the IST (Information Society Technologies) Human Language Technologies Calls from 1999 onwards.
There are three CLASS clusters which address: natural and multimodalinteractivity, cross-lingual knowledge management, and intelligent interactive information presentation. Within each cluster, CLASS focuses on quality enhancement through standards adoption, reference platform/architecture specification, technology exploration and asset sharing, and best development practice. Across clusters, CLASS focuses on evaluation of systems and components including relevant resource issues, on impact and results take-up, and on strongly improved visibility of research.
CLASSiks The Language Technology Lab at DFKI is responsible for the CLASSiks Cluster (Crosslingual Information and Knowledge Management Systems).
Information and knowledge management is one of the fastest-growing application areas for human language technology. Information management is concerned with large volumes of information that modern companies need to gather, produce, exploit and maintain for securing and improving their operations. By far the largest share of the information considered relevant for corporate information management is represented in documents containing texts. The practitioners of corporate information management agree that the efficiency and effectiveness of their work will depend on improved methods for dealing with the contents of corporate information bases. The goal is true content management which can only be automated if IT provides tools exploiting sophisticated knowledge about the structure of human languages. In addition to explicit knowledge, KM is also concerned with the wealth of implicit knowledge that govern the complex processes of an organisation.
At DFKI the CLASSiks Cluster coordination tasks include the specification of a standard reference platform/architecture which, together with the necessary specification of interfaces and tasks, could serve as a base and reference architecture for LT-based improvements in language identification, summarisation, categorisation, retrieval, clustering, relevance ranking, information extraction, information presentation/visualisation, and knowledge discovery. Coordination also includes connecting the involved projects with the I&KM community as well as the promotion of best practice.
- specification of a standard reference platform/architecture which, together with the necessary specification of interfaces and tasks, could serve as a base and reference architecture for LT-based improvements in language identification, summarisation, categorisation, retrieval, clustering, relevance ranking, information extraction, information presentation/visualisation, and knowledge discovery.
- connecting the involved projects with the I&KM community
- promotion of best practice.
