Creating a boxplot from two tables in KNIME

Question

I am trying to plot two data columns coming from different tables using KNIME. I want to plot them in a single plot in order to be easier to compare them. In R this effect can be achieved by the following:

boxplot(df$Delay, df2$Delay,names=c("Column from Table1","Column from Table2"), outline=FALSE)

However, by using KNIME, I cannot think of a way that you can use data coming from two different tables. Have you ever faced this issue in KNIME?

Answer 1

Seems to be a working solution would be -according to the discussions in the comments- the following:

1. Installing the KNIME Interactive R Statistics Integration (you might already have it installed)
2. Using the Add Table To R node to add the second table to R
3. I guess the usual R code can be used to create the figures

Answer 2

If you have the same number of rows in the columns you can use the Column Appender node to get both columns into the same table. If not, you can use the Column Joiner node to create a superset of both columns.

Answer 3

The above answer using the "Add Table to R" node is a very nice option.

You could also do it before hand in KNIME. If the two tables have the same columns you could concatenate them using the "Concatenate" node and, if you need mark the rows with the "Constant Value" node from which table the come from originally.

If the two tables have different columns but some row identifiers in common you could join them using the "Joiner" node into one table. Then pass the concatenated or joined table over to R.