International Journal of Innovations in Engineering and Technology (IJIET)
The real world data today present have missing values that occur due to a variety of reasons. The methods used for data analyzing such as classification, clustering and dimension reduction procedures require complete data. The missing data present in these must be either removed or preferably estimated. Missing data imputation is a key issue in learning from incomplete data. Many techniques had been developed on dealing with missing values in data sets with homogeneous attributes. This paper proposes a higher order spherical kernel based iterative estimator to impute mixed-attribute data sets.