I am really new to KNIME and looking forward to get some help from you. I have the problem, that I have 5 mio rows in CSV and want analyse them. Until now, I alway splitted the data into 1 mio parts. After that I proceed 5 times all the analysis with 1 mio rows, did a PIVOT table and summed the 5 parts together. It took up to 5-6 hours to do that. KNIME was the suggested idea to do it. But I don’t know enough to work with it, unfortunately.
In Excel I used a specific formula to qualify the data and to sort out some rows. Therefore it would be great if you could help me with the right rules.
What kind of rules do I need …
if cell contains “XYZ”, then it is not included
if cell contains “ABC” on position 3-6 in the cell, then it is not included
@DennisMB welcome to the KNIME forum. I have prepared this example on the KNIME hub. You can write a column expression testing a single column for your conditions (0 = a negative condition is met, 1 = the positive condition is met, 2= something else, maybe also a 1?)
Then for the fun of it I have built a system where you could apply the rules to all columns (you want). If we have one negative in a set of columns it is a negative overall and we would need at least one positive mark in all the columns. Other combinations are of course also possible.
Many thanks to you both. I made some experience with the rule based row-filter and with column expressions. I am not far away from reaching my goal. What I need is an expression with AND. Could you please have a look on my expression and tell me where I did a mistake?
My expression was:
if (and(column(“Testing”)=1, column(“A_Validity”)=31/12/2999, column(“B_Validity”)=31/12/2999 )) {1}
else {0}
Error Message is: “Invalid left hand side for assignment”
Oh that worked very well. Thank you. I didn’t know that I have to use == and && instead of = and AND. Additional question: How would the same formula look like in a rule-bases row filter?
