Aspects of object merging

Abstract

Information fusion is a research area that investigates how to combine information provided by independent sources into one piece of information. This topic has been studied for several applications leading to, amongst others, aggregation operators in bounded lattices and merge functions of propositional belief bases. In this paper, information fusion is investigated in the context of coreferent objects, which are objects that refer to the same real world entity. Some important properties of object merge functions are pointed out and object merge functions for both atomic and complex objects are investigated in a possibilistic framework. It is shown how merge functions for complex objects can be composed of merge functions for atomic objects, such that the composite function inherits the properties of the merge functions from which it is composed.