I am attempting to get total counts of values for specific variables within a data set. I originally tried the Groupby node, but was told that the Column Aggregator node would work just fine as I only want to get counts of total number of values for one variable. Each time I configure the Colum Aggregator, I select the Aggregation Column I wish to manipulate and then I select “count” within the options settings. Every time I execute the node, the table returns the same amount of rows I attempted to aggregate and shows clearly matching values as separate values. Every value has a “1” in the count.
Below, I’ve attached two screenshots of the values not matching correctly.
You’ll also see my entire Knime workflow at the bottom.
The variable where I am trying to count values is a variable from the FEC (Federal Elections Commission) that classifies congressional spending into categories. Each number in column 16 is one of those categories.
thanks for providing us with the information and the workflow (unfortunately we cannot read the files which are located on your PC :))! Anyhow, if I understood correctly you want to get a count for all occurrences of i.e. 004 right? For this scenario you should use the Group By Node and not the Column Aggregator. The Column Aggregator in your case counts the columns in your table per row which will always be 1. For two columns you would get Count=2 or for a Max aggregation the highest number per row. To make use of the GroupBy Node you would need to create some sort of dummy column to do a count aggregation on this column.
You could use for instance the Constant Value Column – KNIME Community Hub node and set some arbitrary number. Note you need that in case you only have one column. After that you can group by your “CATEGORY” column and do the aggregation (count) on the dummy column.
@Daniel_Weikert That specific node. It’s somewhat decent at giving directions, but clearly was off the mark with the column aggregator. The value counter node solved my problem.
