Nowadays, it is often that a geographic area is described by several independent geographic databases. Yet users need to fusion various information coming from these databases. In order to integrate databases, redundancy and inconsistency between data should be identified. Many steps are required to finalise the databases integration, in particular automatic data matching. In this paper, one approach of matching geographic data bearing on the belief theory is presented. This approach consists in combining criteria from knowledge such as geometry, orientation, nature of roads, names and topology. Then it is tested on heterogeneous network representing roads.