Missing value imputation
In this section, we will show you how to impute the missing values in the numeric column. Missing values might exist in the raw data, or they can be introduced in the data during the time alignment. We will introduce the missing value imputation in the following subsections:
- Defining different types of missing values
- Introducing missing value imputation techniques
First, we will investigate the different types of missing values.
Defining the different types of missing values
In KNIME, you can recognize a missing value from a red question mark in the data. Furthermore, many nodes, such as the line plot for visualizing time series, provide the option to either remove the missing values or leave them in before performing their tasks. Also, you can inspect the number of missing values in the data in the No. Missing column in the output view of the Statistics node:
Figure 3.7 – Displaying the number of missing...
