Find everything you need to get certified on Fabric—skills challenges, live sessions, exam prep, role guidance, and a 50 percent discount on exams.
Get startedEarn a 50% discount on the DP-600 certification exam by completing the Fabric 30 Days to Learn It challenge.
Hello
I have data fields like:
Gender, Height, Weight, Disease
with values like:
M/F, 110-220cm, 40-250kg, Y/N
How can I identify distinct groups of the above and their probability to have Disease=Y/N, i.e. perform clustering?
I am looking to identify any possible combinations e.g. Weight=120-150kg and Gender=F and Height=150-160cm, can have 80% probability for Disease=Y etc, in addition to single variables i.e. Gender=F alone having 20% probability of Disease=Y.
Also, being able to choose multiple outputs would also be good and of course various types of variables/outputs i.e. numerical, binary, independent or not etc.
Thanks!
Thanks all but I am talking about the statistical machine learning modelling called clustering rather than simply grouping.
Hi @UsePowerBI ,
You could create measure like the below :
test = CALCULATE(COUNTROWS('Table'),ALLSELECTED('Table'),'Table'[Disease]="Y")
Then create visual :
You can choose the range you want to get the results you want by adjusting according to the slicer.
Did I answer your question? Mark my post as a solution!
Best Regards
Lucien
Ben's blog has a piece on clustering:
Clustering in Power BI using R - Ben's Blog (datakuity.com)
Proud to be a Super User!
@UsePowerBI , if all these are columns in same table, you can switch true() and create a new column
Swich( True() ,
[Weight]="120-150kg" && [Gender]="F" && [Height]="150-160cm" , "Y"
//Or you can >= and <= [Height]>=150 && [Height] <=160
// Add others ,
"N") //else value
@UsePowerBI , if all these are columns in same table, you can switch true() and create a new column
Swich( True() ,
[Weight]="120-150kg" && [Gender]="F" && [Height]="150-160cm" , "Y"
//Or you can >= and <= [Height]>=150 && [Height] <=160
// Add others ,
"N") //else value