Ontology-Guided Data Acquisition and Analysis: Using Ontologies for Advanced Statistical Analysis

摘要

We present an ontology-based, domain independent data acquisition and preparation system, which is able to process arbitrarily structured data. The system is fully generic and customizes itself automatically at runtime based on a user-defined ontology. So it can be instantiated for any domain of application by defining an ontology for this domain. Furthermore, semantic rules can be integrated into the ontology to check the data's semantic plausibility. We plan to integrate statistical and data mining algorithms that take advantage of the structural ontology information to allow the user to perform (semi) automatic explorative data analysis. In this paper, the system is described in detail and a motivation for ontology-guided data analysis is given.