Transformation & aggregation — quiz

To build a LEFT JOIN in the Joiner, include:

1. Matching rows only
2. Matching rows and left unmatched
3. All three outputs
4. Right unmatched only

Answer: Matching rows and left unmatched — Left unmatched are the rows with no partner.

Output rows exceed the left input. Cause?

1. A bug
2. Duplicate keys on the right side
3. Missing values
4. Wrong column types

Answer: Duplicate keys on the right side — GroupBy the right branch by the key first.

Concatenating two tables fails on duplicate row keys. Fix:

1. Rename columns
2. Let the node append a suffix or generate new RowIDs
3. Use the Joiner
4. Sort first

Answer: Let the node append a suffix or generate new RowIDs — KNIME requires unique RowIDs.

Column Appender matches rows:

1. On a key column
2. By row position
3. Alphabetically
4. By RowID only

Answer: By row position — Which is wrong unless the tables are genuinely parallel.

After GroupBy, downstream nodes cannot find "Revenue". Why?

1. It was dropped
2. It was renamed to Sum(Revenue) by default
3. Wrong type
4. It became a RowID

Answer: It was renamed to Sum(Revenue) by default — Set naming to keep original names.

Which node gives a running total?

1. GroupBy
2. Moving Aggregation after a Sorter
3. Pivoting
4. Column Appender

Answer: Moving Aggregation after a Sorter — It depends on row order, so sort first.

Which node converts month columns into month rows?

1. Pivoting
2. Unpivoting
3. GroupBy
4. Concatenate

Answer: Unpivoting — Unpivoting is the wide-to-long direction.

Why can a workflow break after a Pivoting node?

1. It is slow
2. New data values create new columns, changing the schema
3. It loses rows
4. It requires sorting

Answer: New data values create new columns, changing the schema — Reference columns by pattern rather than by name.