How to join different examples generated with overlapping start and stoptime

DocMusherDocMusher Member Posts: 333 Unicorn
edited December 2018 in Help

Hi,

Examples have as attributes: Patient ID, LOS (Length of stay), Admissiondate, Dischargedate. 

Some patients have multiple admissions which is normal but some patients have a first Dischargedate being the same as the second Admissiondate resulting in 2 admissions where in reality the patient was only admitted once. How to have 2 examples "joined" in one when in fact one event was timestamped as 2 events. When 2 events are put in one, the Length of stay should be added. 

Thanks

Sven

Answers

  • sgenzersgenzer Administrator, Moderator, Employee, RapidMiner Certified Analyst, Community Manager, Member, University Professor, PM Moderator Posts: 2,959 Community Manager

    hi @SvenVanPoucke - can you Aggregate LOS by Patient ID?

  • DocMusherDocMusher Member Posts: 333 Unicorn

    Aggregation of LOS by patient ID but how to distinguish 2 events which are real (discharge 1 and admission 2) are completely different in time, from 2 events that actually are only one event because discharge 1 and admission 2 are identical or almost identical)

  • MartinLiebigMartinLiebig Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 3,503 RM Data Scientist

    Hi Sven,

     

    Balasz posted on his blog a generic join script: https://datascientist.at/2016/06/generic-joins-in-rapidminer/

    You can also join on date in range and so on.

     

    Cheers,

    Martin

    - Sr. Director Data Solutions, Altair RapidMiner -
    Dortmund, Germany
  • DocMusherDocMusher Member Posts: 333 Unicorn

    Thanks I try this. 

    Sven

Sign In or Register to comment.