Abstract
In this paper we discuss approximation spaces that are useful for studying local lower and upper approximations. Set definability and properties of the approximation space, including best approximations, are considered as well. Finding best approximations is a NP-hard problem. Finally, we present LEM2-like algorithms for determining local lower and upper coverings for a given incomplete data set. Lower and upper approximations, associated with these coverings, are sub-optimal.
Get full access to this article
View all access options for this article.