An efficient use of a mapping knowledge base

In a wide range of commercial and industrial applications data from different sources - web, databases or documents - are integrated into databases for the purpose of querying and analysis. It is then determined for each of the information contained in the source schema, what should be the corresponding column in the target relational database, that is to say, to establish a correspondence (Mapping) between the data source and the target column. A number of tools and prototypes of Schema Matching have been developed in order to automate the process of finding a suitable mapping. One technique employed is to use mapping decisions previously taken by the user manually. These decisions are stored in a Mapping Knowledge Base. The problem with this procedure is that the volume of the Mapping Knowledge Base increases very significantly over time. Processing large volumes of information may take several hours or even days. The prototype described in this article, keeps processing time at acceptable levels.