Hello, I think it would be a great idea to create a MOD for scatterplot matrix. This type of visualization shows the distribution of data per attribute as well as a scatterplot for every pair of attributes.
Today, it is possible to create it using TERR or Python but I still think this is a potential for a MOD.
In other words, you can quickly see how your data distributes as well as any trend of 2 attributes.
Why is important: because this is one of the most basics forms of exploratory data analysis, everyone wants to have a quick glimpse into the data to discover trends and spot collinearity issues before deep diving into the creation of advanced analytic models.
Who would benefit: Everyone exploring data, in any industry.
How it should work: One should have the option to drag and drop a MOD and simply select the attributes and the scatter plot matrix creates itself with the variables placed randomly. They should not be in a specific order, but it would be nice if the person could assign the order or change it in the matrix ay their convenience.
I'm leaving some articles for reference:
https://dzone.com/articles/what-when-amp-how-of-scatterplot-matrix-in-python
https://seaborn.pydata.org/examples/scatterplot_matrix.html
Implemented in | 11.5 |
Scatter plot matrix plot is available as a visualization mod for Spotfire from the TIBCO Community Exchange: https://community.tibco.com/s/exchange/aCv4z000000kPrFCAU/scatter-plot-matrix-mod-for-tibco-spotfire
To use the Scatter Plot Matrix Mod, Spotfire 11.5 or later is required.
Agree. Many customers would use this. We are working to see if we can develop a mod for it.