Ontologies: Dynamic Networks of Formally Represented Meaning

1. SW Portal Internal Research Seminar 04/12/2003 Ontologies: Dynamic Networks of Formally Represented Meaning Dieter Fensel: Ontologies: Dynamic Networks of Formally Represented Meaning, 2001

2. SW Portal Internal Research Seminar 04/12/2003 Introduction - Ontologies: Formal and Real, based on Consensus • Motivation: lack of technology that supports access to unstructured, heterogeneous and distributed information and knowledge sources • Goal: to examine some of the essential requirements for such a technology • WWW – problems: • find, access, present and maintain the information (HTML,SGML,etc.) • information content – natural language => gap between the information available for tools and the information kept in a form legible to humans • computers – devices that post and render information, but no access to the actual content => offer only limited support => hard for humans not only accessing and processing information, but also extracting and interpreting it

3. SW Portal Internal Research Seminar 04/12/2003 Introduction - Ontologies: Formal and Real, based on Consensus • Sematic Web: provides automated information access based on machine processable • semantics of data and heuristics that use these meta data. • the explicit representation of the semantics data + domain theories => a Web that provides a qualitatively new level of service • Ontologies: key enabeling technology for the semantic web. • - need to interweave human understanding of symbols with their machine processability – a closer look to the nature of Ontologies and to the question wheather and how they can actually provide such an service • - developed in AI for knowledge sharing and re-use • very popular, because of what they promis: a shared and common understanding of a domain that can be communicated between people and application systems. • glue together 2 essential aspects: • They define formal semantics for information, allowing information processing by a computer. • They define real-world semantics, which makes it possible to link machine processable content with meaning for humans based on consensual terminologies.

4. SW Portal Internal Research Seminar 04/12/2003 Ontologies – formal semantics • enables machine processing of the semantics of information • formal semantics can be achieved by a layered language architecture • The onion model to control complexity

5. SW Portal Internal Research Seminar 04/12/2003 Ontologies – real-world semantics • how can Ontologies be used to communicate real-world semantics between human and artificial agents? • difference between viewing ontologies as “true” models of the real world or steps in a process of organizing evolving consensus. • ontologies can only be viewed as a network of interwoven ontologies(may have overlapping and excluding pieces, and must be dynamic in nature) => Ontologies as • dynamic networks of formally represented meaning • ontologies must have a network architecture and Ontologies must be dynamic. Joined set of symbols and a consensual interpretation Agent 1 Agent 2 Communication

6. SW Portal Internal Research Seminar 04/12/2003 Heterogeneity in Space: Ontology as Networks of Meaning (1) • an island of meaning must be interwoven to form more complex structures enabling exchange of information beyond domain, task, and sociological boundaries. • tool support must be provided to define local domain models that express a commitment of a group of agents that share a certain domain and task and that can agree on a joined world view for this purpose. • links must be defined between these Ontologies and this network must allow overlapping Ontologies with conflicting, and even contradictory - conceptualizations. • heterogeneity has been an essential requirement for this Ontology network. • ex: Gnutella (a P2P network) – agents were able to enter and leave the network dynamically; they could also communicate with a local environment of other agents.

7. SW Portal Internal Research Seminar 04/12/2003 Heterogeneity in Space: Ontology as Networks of Meaning (2) • What is needed is focus on: • • linking local conceptualizations that deal with heterogen definitions and personalized views. • • support in easy configuration and re-configuration of such networks according to the communication needs of agent coalitions. • • methods and tools that help agents to organize consensus, allowing them to exchange meaning.

8. SW Portal Internal Research Seminar 04/12/2003 Development in time: Living Ontologies • ontologies as pre-requisite for consensus • • ontologies as a result of consensus • an ontology is as much required for the exchange of meaning as the exchange of meaning may influence and modify an ontology => evolving ontologies describe a process rather than a static model. • the real challenge: protocols for the process of evolving ontologie. • evolving over time - an essential requirement for useful ontologies. ontologies cannot be understood as a static model

9. SW Portal Internal Research Seminar 04/12/2003 Conclusions • ontologies help to establish consensual terminologies that make sense to both sites. • • computers are able to process information based on their machine-processable semantics; humans are able to make sense of this information based on their connection to real-world semantics. • • a model or “protocol” for driving the network that maintains the process of evolving ontologies is the real challenge for making the semantic web reality. • Ontologies as Networks of Meaning and Living Ontologies