Optimization of Missing Value Imputation using Reinforcement Programming

Missing value imputation is a crucial and challenging research topic in data mining because the data in real life are often contains missing value. The incorrect way to handle missing value will lead major problem in data mining processing to produce a new knowledge. One technique to solve Missing value imputation is by using machine learning algorithm. In this paper, we will present a new approach for missing data imputation using Reinforcement Programming to deal with incomplete data by filling the incompleteness data with considering exploration and exploitation of its environment to learn the data pattern. The experimental result demonstrates that Reinforcement Programming runs well and has a great result of SSE of new data with assigned value and shows effectiveness computational time than the other five imputation methods used as benchmark.