NCSR "Demokritos"

Institute of Informatics and Telecommunications

Software & Knowledge Engineering Laboratory

NCSR "Demokritos" is the largest self-governing research organisation, under the supervision of the Greek Government. NCSR participates in MedIEQ with the Software and Knowledge Engineering Laboratory (SKEL) of its Institute of Informatics and Telecommunications (http://www.iit.demokritos.gr/skel [1]).

SKEL work focuses on intelligent systems and services for the Web and the Semantic Web, and more specifically: web content collection and extraction for specific thematic areas, knowledge discovery from web usage data, creation and maintenance of ontologies to support the collection and extraction tasks, natural interaction with web-based information systems.

SKEL work is structured around the following major R&D activities:

• Information Integration. Research and development in the areas of information filtering and extraction from web content and raw text, knowledge discovery from text or data bases. This activity exploits and develops methods, techniques and tools from machine learning, language and knowledge engineering. At the level of the infrastructure, the emphasis is on tools to facilitate the development of new applications as well as their customization to new domains and languages. In the context of this activity the spin off private company, i-sieve technologies ltd., was set up. Its main focus is to exploit commercially the in-house developed technology for online multimedia content analysis (http://www.i-sieve.com [2])

• User-friendly Information Access. Research and development in: (a)personalization of web-based information services, through the analysis of user behaviour, (b)dialogue systems which enable the communication of the user with computers or other devices in natural language (spoken or written dialogues). This activity exploits and develops methods, techniques and tools from language engineering, user modelling, software engineering and machine learning. At the level of the infrastructure, the emphasis is on a generic personalization server and authoring tools to support dialogue systems.

• Ontology Creation and Maintenance. Research and development in (a)ontology design, (b)ontology population and enrichment, (c)use of ontologies for the semantic annotation of content. This activity exploits and develops methods, techniques and tools from machine learning, knowledge representation, and language engineering. At the level of the infrastructure, the emphasis is on tools for ontology creation and maintenance which are based on general purpose platforms for ontology and language engineering.

During 2024-2005, SKEL coordinated or participated in various EC and national funded projects. SKEL researchers also initiated the cooperation with research groups outside Europe (USA, Canada, Singapore, Japan) which so far involves the exchange of research results and the issue of common publications, but which may lead to also funded projects in the near future. During this period, SKEL researchers published their research results in a large number of papers in international scientific journals, and in the proceedings of international conferences and workshops.

SKEL strategy for research emphasises on the development of the necessary infrastructure to support research activities and the development of intelligent systems and services. This effort has resulted in a variety of tools and resources, some of which are available to and are being used by the international research community:

• The multi-lingual, cross-platform, general-purpose text engineering environment, Ellogon (http://www.ellogon.org/ [3]) which is used by several research teams and companies internationally.

• Language processing tools (morphological, syntactic analysers) and linguistic resources (morphological lexicon, grammars) for the Greek language. The Greek morphological lexicon of SKEL is used, under license, by research groups in Europe.

• A platform for developing web content collection and extraction systems.

• The FILTRON e-mail filter, that blocks unsolicited commercial e-mail (spam messages), and is tailored automatically to the incoming messages of the inbox it protects by analyzing previously received messages.

• The FilterX Web proxy filter, that blocks obscene Web content using text and image information at real-time.

• The PServer general-purpose server for personalization, incorporating our state-of-the-art user modeling technologies.

• An authoring tool for porting language generation systems to new domains and languages.

• Tools for creating and maintaining ontologies which are based on the Protégé ontology engineering platform, the ontology management system developed in the context of the FP5-IST project CROSSMARC and the language engineering platform Ellogon.

• The KOINOTHTES system for knowledge discovery from web usage data.