BrownBag - 23.10.2008 - Towards Service-oriented Knowledge Discovery with Ontologies

Nada Lavrac, Vid Podpecan, Monika Zakova, Pance Panov:
Towards Service-oriented Knowledge Discovery with Ontologies

A major challenge for third generation data mining and knowledge discovery systems is the integration of highly diverse data/knowledge resources and tools for data integration, data mining and knowledge discovery which are distributed across the network. Third generation data mining and knowledge discovery will be made possible through semantic annotations of different information resources and data mining & knowledge discovery systems, and by implementing data mining and knowledge discovery tools as services available on the web. Such service-oriented data mining and knowledge discovery systems will enable meta-level search of data/knowledge resources and systems, enabling the construction of knowledge discovery workflows (potentially repeatable sequences of data mining and data integration steps) resulting in improved pattern and model discovery.

The brown bag seminar will describe the ongoing joint work of the Ljubljana, Prague and Leiden data mining groups in the development of a prototype environment for service-oriented knowledge discovery, enabling automated data mining workflow construction using a data mining ontology. The presentation and discussions will be structured as follows:
- Nada: Introduction to the semantic data mining framework for service-oriented knowledge discovery
- Vid: Data mining services
- Pance: Data Mining ontology
- Monika: Automated DM workflow construction using the CTU DM ontology
- All: Current work of Ljubljana, Leiden and CTU
- All: Future joint developments
- All: Discussion