A Novel Algorithm for Filling Incomplete Data Based on Attribute Reduction

Traditional data filling algorithms take the unified method to fill the incomplete data, leading to the low efficiency for filling a large number of incomplete data. This paper proposes a novel algorithm for filling incomplete data based on attribute reduction. The presented algorithm first uses the attribute reduction to distinguish the important attribute from unimportant attribute, and then fills the two kinds of data attributes using different filling technologies. Experiments show that this algorithm can effectively and quickly fill the large-scale incomplete data. Keywords-attribute reduction; incomplete data; data filling