Designing medical production rules from semantically integrated data

In the paper an algorithm for automatic knowledge acquisition is proposed. The knowledge is acquired from aggregate data stored in different repositories. The algorithm operates by means of semantic data integration, allowing both syntax and semantic differences between data coming from different sources. If only we know data taxonomies, can interpret data schemas and design schema mappings, then the differences are not an obstacle to integration. The acquired knowledge is being defined in a form of production rules with uncertainty. The considerations are illustrated with medical examples.