I am working on a project of integrating the preferences of the workers into a schedule, I mean we won’t only satisfy the systematic constraints but also users preferences as constraints, so we are trying to learn these preferences from historical data. We are focusing on absences. I tried to model this problem as a graph and then embed it so I use later an LLM to generate these preferences and transform them into constraints. I need any ideas on how to present the data as graph. The first idea I used is a tree of 3 levels : day, period and absence cause, the data I have is of this type (employeeid, sex, age, contract-start-date, absence-type, start-date of absence, end date, duration). Any ideas ? I need many graphs representations because i am willing to use an ensemble learning technic on the resulting embeddings so I can get a good representation.
Asked
Active
Viewed 11 times
