This KNIME workflow demonstrates usage of the Mahalanobis Distance calculation. The calculation is done in two steps- first the distance metrics is calculated and then it is used to calculated actual distances using Distance Matrix Calculate node. The distances are calculated between 3 points using a different sample to define covariance: A 0.3 1.1 B 0.45 0.8 C 0.45 1.4 Note, that the Euclidian distance between points (A,B) is the same as (A,C). However, the calculated Mahalanobis distance is much larger for (A,B)- 5.50 vs 0.96, because the variance in that direction is much smaller. The workflow also shows how to split the collection of distances to get representation, that could be dumped into a scv file.
This is a companion discussion topic for the original entry at https://kni.me/w/c1ORmfZL4sXyx-Xn