In the previous note, I have shown an example for usage of “corrplot” package to visualize the correlations between different objects. Another way is using a heatmap table. Especially if the measure of similarity/dissimilarity is not correlation and the distribution of the values is skewed, using heatmap table offers much more flexibility. What is the difference between a heatmap and a heatmap table? Well, cannot say there is much difference. Heatmap is basically a table where the values are represented by the colors and a good way to summarize a large volume of data. Heatmap table can be thought of as a small version of a heatmap where you can also add the values. Basically it is a coloured table :) Enough of words, let’s see what it is and when it is useful: check the example here.
There are several possible ways to visualise correlation matrices in R. I think one of the most common ways is using a heatmap. However, in R there is a package called “corrplot” designed specifically for this purpose. You can install this package from CRAN using: