This KNIME component allows you to apply various data cleaning steps interactively. Default configuration will implement cleaning of missing values and outliers.
Availble pre-processing steps:
- Automatic type guessing: determine the most specific type in each string column and change the column types accordingly.
- Treatment of missing values: separate configurations for missing values in string and number columns.
- Outlier removal: configuration on how to treat outliers.
All preprocessing steps have a "Do nothing" option, that will allow to skip the corresponding step.
In addition, you can select and (optionally) rename a subset of columns to be used in the following. If no columns are selected, whole data table will be used.
There is also a summary of the input data presented at the bottom of the page to drive decisions about pre-processing steps.
Use WebPortal/JS View to configure this components
Nodes
- Text Output Widget7 ×
- Transpose6 ×
- Table Row to Variable5 ×
- CASE Switch Data (End)4 ×
- CASE Switch Data (Start)4 ×
- Show all 43 nodes
Extensions
- FeatureKNIME Base nodes
- FeatureKNIME Column Expressions (legacy)
- FeatureKNIME JavaScript Views
- FeatureKNIME JavaScript Views (Labs)
- FeatureKNIME Javasnippet
- Show all 8 modules
Links
- No links available