Analysis of Missing Value Imputation Application with K-Nearest Neighbor (K-NN) Algorithm in Dataset

Abstract

Missing value is a problem that is still often found in many studies. Missing value is where data or data features are not available completely and intact. This still happens a lot in datasets that will be used in research. The missing value is caused by many factors such as human error, unavailable data or even from a virus in the database. Data is important for research, incomplete data will affect the results obtained. Data mining is a process that is very influential on data, including the classification process. Classification in data mining can be done if the data is complete. These problems can be overcome by the Imputation process by combining it with the K-Nearest Neighbor process or the process can be called K-Nearest Neighbor Imputation (K-NNI). In the research that has been done the K-Nearest Neighbor Imputation algorithm can overcome the problem of missing values in the dataset. This can be seen from the level of accuracy obtained where the accuracy of the classification process before handling the missing value is 77.01% while after the imputation process the accuracy is 78.31%