An Effective Imputation Technique for Improving the Performance of Skyline Queries for Incomplete Database

Skyline query is one of the types of user preference-based query which is used to retrieve non-dominated results. This is extensively used in applications like recommender system, decision-making applications and e-commerce. Existing traditional skyline algorithms are well suited only for complete data. However, we find in the real time dataset we come across one or more missing data and this will lead to incorrect results during skyline query processing. Due to which the user may not get accurate information while they give some preferences. In this paper, we propose the Multiple Prediction based Imputation Algorithm (MPIA)for incomplete data and finally execute the skyline query which gives better results than the existing approaches like Mean, Median, K- Nearest Neighbor (KNN)and Multiple Imputation Chained Equations (MICE).