In order to fully benefit from the three core technologies in SEKT: Ontology-based Metadata, Human Language Technology and Knowledge Discovery, they must be used together. This convergence is now timely because of the maturing of the three separate disciplines, particularly ontology technology, which has received much attention over the last 2-3 years. Thus, in SEKT, the ontology learning software based on knowledge discovery techniques will develop ontologies which will be populated with metadata using software employing human language technology. These ontologies and their corresponding metadata will be managed and evolved using enhanced ontology and metadata technology developed in SEKT. In turn, the ontology evolution software will benefit from knowledge discovery techniques. Inferencing technology will be developed to inference over the ontologies and mediation software will be used to align, merge and translate between different but related ontologies.

• Ontology & Metadata Technology

Next Generation Knowledge Management solutions will be built upon ontology-based metadata (OMT) and thus the creation and management of machine-interpretable information and the consequent use of ontologies. The integrated management of ontologies and metadata, especially their generation, mediation and evolution, is fundamental to this development and relies in part on innovative Human Language Technology (HLT) and Knowledge Discovery (KD) methods.

Advanced reasoning capabilities will strongly support the evolution of ontologies and metadata and greatly reduce the overhead for maintenance. Work in advanced reasoning will include the development of techniques for robust reasoning, i.e. reasoning in the presence of inconsistencies, i.e. in order to give meaningful results even when the overall ontology has conflicts. It will also include flexible reasoning which can cope with changes and conflicts in a given model and can fall back to old versions or change the scope of reasoning to a consistent set of statements. The advanced reasoning work will support the evolution of ontologies and meta-data, in order to reduce maintenance overhead.

• Human Language Technology

In this project Human Language Technology (HLT) has two key contributions that will both include the handling of multilinguality. In the first place it will be used to semantically annotate informal and unstructured knowledge. Thus, the automatic or semi-automatic extraction of metadata from legacy data will be achieved. Secondly natural language processing will be used to generate natural language based on formal knowledge (ontologies and metadata). Here, HLT will be strongly integrated with methods from KD and OMT. The ontologies that structure metadata are in many cases language-independent to a significant degree. SEKT will trial metadata generation methods based on Information Extraction, Content Extraction and other language analysis technology that is used in HLT for various languages and, similarly, prove the documentation of ontologies and metadata in practice (using Natural Language Generation).

All the HLT technology used and further developed in the project will be based on systems that have been proven in a large range of languages. For example, the extraction components were recently entered in the “TIDES Surprise Language Competition” (that measured the ability to port HLT systems to Hindi) with favourable results. All the main European languages, and others from Chinese to Bulgarian, have also been covered in various ways, and SEKT will include at least four languages directly in the case studies (English, German, French and Spanish).

• Knowledge Discovery

Knowledge discovery (KD) is also a key technology in its own right for knowledge management, helping to provide knowledge workers with the knowledge they need from large volumes of data. At the same time, like natural language processing, knowledge discovery has a role in metadata extraction to enable the automatic or semi-automatic mark-up of data, and in ontology learning and evolution. Obviously, KD interacts closely with HLT for e.g. metadata generation, and ontologies provide the background knowledge for improving KD results.