From ViRBO

Contents Semantic Web Applications User Interface Frameworks (e.g., faceted Search) Semantically Assisted Discovery Linked data Provenance Application Support The Big Issues Where is the linking??? Apps working with each other through shared/linked ontologies? Why so difficult??? Performance of large triple stores Linking / Sharing Ontologies Clarity Coherence (internal consistency) Extendibility Minimal Encoding bias Minimal ontological commitment

1. Semantic Web Applications

1.1. User Interface Frameworks (e.g., faceted Search)

• Earth System Grid

1.2. Semantically Assisted Discovery

• e.g., Noesis

1.3. Linked data

1.4. Provenance

• e.g., Multi-Sensor Data Synergy Advisor

1.5. Application Support

• e.g., Data Quality Screening Service (DQSS)

2. The Big Issues

2.1. Where is the linking???

2.1.1. Apps working with each other through shared/linked ontologies?

2.1.2. Why so difficult???

=== How do get domain knowledge out of the expert's head and into a model?

2.2. Performance of large triple stores

3. Linking / Sharing Ontologies

• From Gruber's [1], Toward Principles for the Design of Ontologies Used for Knowledge Sharing (1993)

3.1. Clarity

3.2. Coherence (internal consistency)

3.3. Extendibility

3.4. Minimal Encoding bias

• bias results when representation choices are made for convenience of notation or implementation

3.5. Minimal ontological commitment

• make as few claims as possible about the world being modeled