Distance matrix to coordinates

I would like to know how to convert the “nearest label” and “distance values” (obtained with “Similarity Search”) into Coordinates to visualize with Scatter Plot.

image

Hi,
I don’t understand how you want to infer coordinates from that data. If you want to lay out the data points using the distances, you need a symmetric distance matrix for multi-dimensional scaling. As far as I know there is only such a node for molecules, though (see Multi-dimensional Scaling). Otherwise, the scatterplot in KNIME also allows nominal columns for the axes.
Kind regards
Alexander

Hi Alexander

Yes exactly, what I refer to is multidimensional scaling. I would like to generate a plot view similar to the attached picture.

2020-03-17_16h34_40

Hi,
can you use a Python Snippet node? This would allow you to use for example scikit-learn: https://scikit-learn.org/stable/modules/generated/sklearn.manifold.MDS.html. You can also do it in R using the R Snippet: https://www.statmethods.net/advstats/mds.html. Unfortunately there is no native KNIME node to do this.
Kind regards
Alexander

This topic was automatically closed 182 days after the last reply. New replies are no longer allowed.