I have a dataset which has an
interaction_id column and
| interaction_id | events | | -------------- | -------------- | | i123 | enquiry | | i123 | sell | | i123 | update details | | i978 | view options | | i999 | sell | | i999 | enquiry |
As you can see, a unique
interaction_id can have multiple
events that belongs to it. The problem that I am trying to solve (or to validate if possible or not to solve) is to structure the above data so it can answer these:
- when there is an
sellevent, what other events are there?
- how many interactions have both
At the moment, the solution I have thought about is adding tagging columns for each
interaction_id but this solution can only answer the 2nd question above.
Do you have any suggestion on how to tackle this problem? Or is it simply unable to be solved using data structure?