Missing Value Imputation Methods and Algorithms Evaluation

Data mining has achieved spectacular success in practically every sector, including finance and banking, retail sector, insurance and healthcare, scientific analysis, telecommunication industry, research and so on. Since, real-world data are highly susceptible to noisy, missing, and inconsistent data due to their typically huge size and their likely origin from multiple, heterogenous sources. Low-quality data will lead to low-quality mining results. To improve the quality of data number of data preprocessing techniques are applied to clean the data and as consequence data mining results. One of the major irritating issues with real world data is missing values (MVs). Attributing missing values of data improves classification accuracy. This paper presents many methods and techniques for dealing with missing data. The paper also sheds insight on potential limitations and research needs. According to the results of a survey, the most commonly utilized algorithm performance indicators are assumptions, accuracy and time complexity.