The following node is available in the Open Source KNIME predictive analytics and data mining platform version 2.7.1. Discover over 1000 other nodes, as well as enterprise functionality at
http://knime.com.
Low Variance Filter
Filters out double-compatible columns, whose variance is below
a user defined threshold. Columns with low variance are likely to
distract certain learning algorithms (in particular those which are
distance based) and are therefore better removed.
Note, the input table should not be normalized.
Dialog Options
Column Filter
- Variance Upper Bound
-
Choose a variance value here. The higher the value, the more columns
are likely to get filtered out. Choose 0 to filter for columns, which
only contain one constant value.
- Include
-
This list contains the column names that are considered for
filtering. Any other column will be left untouched (i.e. will also be present
in the output table independent of their variance).
- Enforce Inclusion
-
Select this option to enforce the current inclusion list to stay
the same even if the input table specification changes.
New columns will automatically be added to the exclusion list.
- Select
-
Use these buttons to move columns between the Include and Exclude
list.
- Search
-
Use one of these fields to search either within the Include
or Exclude list for certain column names or name substrings.
Repeated clicking of the search button marks the next column that
matches the search text. The check box 'Mark all search hits' causes
all matching columns to be selected making them movable between
the two lists.
- Exclude
-
This list contains the column names of the input table that
will be left untouched (i.e. will also be present
in the output table independent of their variance).
- Enforce Exclusion
-
Select this option to enforce the current exclusion list to stay
the same even if the input table specification changes.
New columns will automatically be added to the inclusion list.
Ports
Input Ports
0 |
Numeric input data. (Non-numeric columns will be left untouched.) |
This node is contained in KNIME Base Nodes
provided by KNIME GmbH, Konstanz, Germany.