Forum Discussion
Key Influencers Modelling in Power BI
Hello community,
I made use of the 'Key Influencers' visualisation on Power BI to explain the impact of factors influencing my KPI output. Have a few questions that need help:
1. Just to satisfy my curiosity, does anyone knows what sort of modelling algorithm is 'Key Influencers' using? Is it linear/exponential/ridged regression based modelling? Or is it something more complicated?
2. My data is categorical, say has a scale of 'Agree-Disagree' from survey. Does this mean that the model is saying when a particular factor has 1 additional person saying 'Agree', the likelihood of driving satisfaction goes up by 3.36x? Satisfaction is being measured in %.
Thanks!
#1 is a question for Microsoft's product team
#2 says that if a user agrees with "friendly and courteous community" then that user is 3.36x more likely to be satisfied with the CSAT than if that user disagrees with "friendly and courteous community".
2 Replies
#1 is a question for Microsoft's product team
#2 says that if a user agrees with "friendly and courteous community" then that user is 3.36x more likely to be satisfied with the CSAT than if that user disagrees with "friendly and courteous community".