Information-decomposition-model-based missing value estimation for not missing at random dataset

Missing data estimation is an important strategy
for improving learning performance in learning from
incomplete data, especially, when there are non discardable
records with missing values. However, most of the existing
algorithms are focused on missing at random (MAR) or
missing completely at random (MCAR), and less attention
has been paid to data not missing at random (NMAR). In
this paper, an information decomposition imputation
(IDIM) algorithm using fuzzy membership function is
proposed for addressing the missing value problem under
NMAR. Firstly, the proposed IDIM algorithm is presented
with detailed examples. Then, the proposed approach is
evaluated with extensive experiments compared with some
typical algorithms. The experimental results demonstrate
that the proposed algorithm has higher accuracy than the
exiting imputation approaches in terms of normal root
mean square error (NRMSE) and TP?TN evaluation under
different missing strategies.