The application scenario ORDO is intended to research and develop semantic technology, thereby creating new services and software tools that will enable users to organize their entire range of digital information. As the organization process will be automated, it will also be transparent, and require no effort on behalf of the user. High scalability results in problem-free processing of extremely large quantities of data and graphic visualization of that data in the form of knowledge models. Unlike solutions in use today, this personalized linking allows unstructured and structured data to be organized in a uniform manner, making efficient, individual knowledge management possible.
Part of the ORDO application scenario is the development of innovative procedures for the automatic recognition and comprehension of content. This will allow for the recognition of concepts and their respective context or mutual relationships, the extraction of facts and the creation of automatic synopses or summaries. These capabilities are a major improvement on classic abstraction services and offer scope for much more efficient information management. The second step will be to explore suitable methods of extracting information and applying the same advantages to multimedia content.
Research work will also focus on developing efficient procedures for the semi-automatic creation or refinement of ontologies using progressive, linguistic and statistical methods. Development of classification technology for highly scalable, automated classification and clustering (automatic recognition of frequently recurring topics or subject matter) of information is also planned.
The pilot CAPRI, which is aimed primarily at R&D departments in the chemical and pharmaceutical industries, should be one important result of the work of this application scenario. Its main focus will be on finding practical solutions to the problems caused by the ever-increasing amount of data and information that professional scientific knowledge managers are faced with on a daily basis. Conventional search and retrieval technology is unable to penetrate these primary data bases, as many information bases of this kind are not yet accessible via the internet (Deep web). In addition, secondary information sources, such as research reports and other scientific publications, must be evaluated separately. This also applies to active and archived emails, in-house wikis and blogs. The goal for CAPRI research is to develop methods and technologies that allow primary databases to be utilized and combined with internet content. The next step will be to transfer the knowledge acquired from the CAPRI pilot project to the pilot application MEMEX, where it will be developed further to meet the needs of private users. The ultimate aim is to bring digital order to desktop PCs everywhere.