Incremental Three-Way Decisions with Incomplete Information
The theory of rough sets proposed a framework for approximating concepts by three pair-wise disjoint regions, namely, the positive, boundary and negative regions. Rules generated by the three regions form three-way decision rules, which are acceptance, deferment and rejection decisions. The periodic updating of decision rules is required due to the dynamic nature of decision systems. Incremental learning technique is an effective way to solve the problem of dynamic data, which is capable of updating the learning results incrementally without recalculation in the total data set from scratch. In this paper, we present a methodology for incremental updating three-way decisions with incomplete information when the object set varies through the time.
Keywordsthree-way decisions incremental updating incomplete decision system
Unable to display preview. Download preview PDF.