Data and information play an important role in all aspects of science, so data must be processed well through the process of data excavation or data mining. The excavation of patterns from data can be done using machine learning algorithms such as linear regression. However, in the process of extracting information from data, it can be less effective if there is a loss of value in a data. The purpose of this research is to implement the mean imputation and single center imputation chained equation (SICE) techniques against the linear regression algorithm. The data used in this research is numerical data. The root mean squared error (RMSE) value shows that the implementation of linear regression algorithm using the mean imputation technique results in better performance compared to the SICE imputation technique.