Mapping Language for Information Integration Haridimos Kondylakis 1 , Martin Doerr 1 , Dimitris Plexousakis 1 1 Institute of Computer Science, FORTH-ICS P.O. Box 1385, GR 71110, Heraklion, Crete, Greece {kondylak, martin, dp}@ics.forth.gr Technical Report 385, ICS-FORTH, December 2006 Abstract. An essential matter in heterogeneous database integration is the mapping process. In this report we present a mapping language for information integration under a common knowledge representation model (LAV approach). Based on particular requirements for a sufficient quality of information integra- tion to be achieved, we present the most common cases of heterogeneity en- countered in a wide sample of cases from museum collections, archeological, medical and genomic data and reflect the individual investigated domain. We propose a specific mapping annotation format that is capable to capture all those cases. We assume that the level of detail of this format is sufficient to produce complete mediation of data transformation algorithms without further input from the domain experts. This assumption has to be verified in further work. 1. Introduction Data Integration is one of the key problems for the development of modern infor- mation systems. The exponential growth of the web and the extended use of database management systems has brought to the fore the need for seamless interconnection of diverse and large numbers of information sources. In order to provide uniform access to heterogeneous and autonomous data sources, complex query and integration mechanisms have to be designed and implemented. Fig. 1.: The basic mapping schema An essential matter in heterogeneous database integration is the mapping process. We define the mapping of two schemata as a sufficient specification to transformation of each instance of schema 1 into an instance of schema 2 with the same meaning as shown in figure 1. The definition should be independent of particular instances. Map-