Discovering patterns of missing data in survey databases: An application of rough sets

Databases for data mining often have missing values. Missing data are often mistreated in data mining and valuable knowledge related to missing data is often overlooked. This study discusses patterns of missing data in survey databases. It proposes a framework of rough set rule induction method that enables the data miner to obtain association rules of patterns of missing data in a survey database. Through an experiment on a real-world data set, we demonstrate the approach to discovering knowledge about missing data.