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.

Stresser

Produces stress for the given values. Each cell is stressed individually with the given percentages of stress. There are four kinds of stress, based on the column type.

String columns can be stressed by switching the string value to another value inside the given column domain. Otherwise they can be stressed by exchanging two letters.

Number values can be stressed inside or outside the input range.

Dialog Options

Percentage to stress per Row
the percentage to stress. The rows are selected based on this probability.
Percentage to stress per Column
the percentage to stress a column with. If the row was selected to be stressed, the algorithm chooses which cells should be stressed based on this value. If all columns in the stressed row should be changed, set the value to 1. The probability a cell is stressed is the multiplication of both, the row and column stress probability.
String
the type of string stressing applied to all selected string columns
Number
the type of number stressing applied to all selected number columns
Seed
the random seed to produce a deterministic result
Column Selection
Move the columns to be stressed into the "Include" list

Ports

Input Ports
0 Arbitrary input data.
Output Ports
0 Input data with some stressed values.
1 Containing only changed values. Unchanged values in rows which changed will set as missing values.
This node is contained in KNIME Datageneration provided by KNIME GmbH, Konstanz, Germany.