Java Distance


Enables the definition of arbitrary distance measures using a Java snippet. A distance measure is defined on two equal structured rows. You can refer to a cell value by enclosing the column name prefixed by either row1 or row2 for example $row1.c1$ .


Compute the Euclidean distance of the columns c1 and c2 :

Compute the arithmetic mean of the Levenshtein distance of the columns c1 and c2 :
mean(levenshtein($row1.c0$,$row2.c0$),levenshtein($row1.c1$, $row2.c1$))

Note that strings which are part of the expression and are not from the input data (or the result of another wrapped function call) need to be enclosed in double quotes ('"'). Additionally, if the string contains a quote character, it must be escaped using a backslash character ('\"'). Finally, other special characters such as single quotes and backslashes need to be escaped using a backslash. For instance, a single backslash in a string is written as two consecutive backslash characters; the first one acts as the escape character for the second.

Input Ports

  1. Type: Data Input table.

Output Ports

  1. Type: Distance Measure The defined distance.

Find here

Analytics > Distance Calculation > Distance Functions

Make sure to have this extension installed:

KNIME Distance Matrix

Update site for KNIME Analytics Platform 3.7:
KNIME Analytics Platform 3.7 Update Site

How to install extensions