A Sketch-based Conceptual Level Data Integration Methodology

Nowadays, enterprises often need to integrate heterogeneous data sources for many purposes. Given the in- herent complexity of this process, several solutions based on visual manipulation paradigms have been proposed to shoot at the complexity of low-level data integration activities. However, when dealing with complex data sources the use of visual notations might suffer from scale-up problems. One way to solve this problem is to raise the data integration activity at more an abstract level. In this paper, we present a methodology for reconciling data sources at the level of their conceptual schemas. The methodology provides a sketch-based language for manipulating conceptual schemas, a tool prototype, and logic based inference mechanisms to handle lower level schema integration, metadata generation, and loading of data from sources to the reconciled database. We also provide a usability study to prove the effectiveness of the proposed methodology.