Deviation Approach to Missing Attribute values in Data Mining

Abstract

In real-life data, information is not complete because of presence of missing values in attributes. Several models have been developed to overcome the drawbacks produced by missing values in data mining tasks. Statistical methods and techniques may be applied to change an incomplete information system to a complete one in preprocessing/imputation stage of Data Mining. With the help of statistical methods and techniques, we can recover incompleteness of missing data and reduce ambiguities. In this work, we introduce a mean deviation method by which missing attribute values may be replaced with minimum computational complexity when they occur at random.  Keywords: Data Mining, Missing attribute Values, preprocessing, Incomplete Information, Deviation approach.Deviation Approach to Missing Attribute values in Data Mining