Dealing with Missing Data using a Selection Algorithm on Rough Sets

Abstract

This paper discusses the so-called missing data problem, i.e. the problem of imputing missing values in information systems. A new algorithm, called the ARSI algorithm, is proposed to address the imputation problem of missing values on categorical databases using the framework of rough set theory. This algorithm can be seen as a refinement of the ROUSTIDA algorithm and combines the approach of a generalized non-symmetric similarity relation with a generalized discernibility matrix to predict the missing values on incomplete information systems. Computational experiments show that the proposed algorithm is as efficient and competitive as other imputation algorithms.

Missing Data Imputation Approach Based on Adaptive Compressed Sensing

“…This method can directly process multiple missing values, however, it performs poorly when running on large data sets [4]. Rough set theory is an effective way to deal with the problem of uncertainty, on the other hand, it cannot complete parameter optimization and missing data classification, which results in low precision of data completion [5]. Random forest (RF) can process high-dimensional data with high accuracy, while it has a large computational cost when processing a large amount of data [6].…”