Topological solution of missing attribute values problem in incomplete information tables

In this paper, we present a method of data decomposition to avoid the necessity of reasoning on data with missing attribute values. We defined a general binary relation on the original incomplete data which decomposed into data subsets without missing values. Next, topological base relation method for classifier induction is applied to such sets. New approach to find the missing values in consistent data is discussed. New pretopological approximations of concepts are studied and some of their properties are proved. Pre-topological measures are initiated in incomplete information systems and more pre-approximations are defined and studied. Finally, the reducts and core are determined.