I have an excel file in which I have three columns. The first one is the name of an event, the second one is the moment the event starts and the third one is the time at which an event ends. Let's say that I am dealing with times at which people log into a server.
Some rows would be:
Robert 2023-11-28 13:34:43 2023-11-28 13:34:45
Alice 2023-11-28 13:35:22 2023-11-28 13:37:01
Robert 2023-11-28 13:44:43 2023-11-28 13:54:08
Anna 2023-11-28 13:45:02 2023-11-28 13:47:04
My idea is to generate another file using python with as many columns as people and as many rows as seconds and put a 1 if the person is logged in said second and a 0 otherwise. Then, use Hierarchical cluster with average linkage and Roger's Tanimoto distance on the columns to see the closeness between people, and then use association rules considering each row as a basket. I can also calcularte the correlation matrix between people.
I would also like to know about other people's inputs in this topic.
In the end I want to create scenarios based in which people are online at the same moment, e. g. Robert and Samantha are generally online at the same time.
Any input on how to do this is appreciated.
