1. Somethin is "on average larger/smaller " than what? 2. How to explain "on average"?on average of what? 3. How to calculate the percentage numbers? what is the formula?
sgenzerAdministrator, Moderator, Employee, RapidMiner Certified Analyst, Community Manager, Member, University Professor, PM ModeratorPosts: 2,959 Community Manager
IngoRMAdministrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, Community Manager, RMResearcher, Member, University ProfessorPosts: 1,751 RM Founder
Hi,
When it says "X is on average Y% higher" here is what it means:
The average value for feature / attribute X for the examples / cases in the cluster is Y% higher than the average values of feature X for the examples which are not in the cluster.
BTW, those statements are a textual summary of the top 3 contributors as shown in the Heat Map chart.
Try to put your data into a process like this. As you can see I added an aggregation step to the automodel results so we can see some additional averages.
Answers
2. How to explain "on average"?on average of what?
3. How to calculate the percentage numbers? what is the formula?
tagging @IngoRM
Scott
Hi,
When it says "X is on average Y% higher" here is what it means:
The average value for feature / attribute X for the examples / cases in the cluster is Y% higher than the average values of feature X for the examples which are not in the cluster.
BTW, those statements are a textual summary of the top 3 contributors as shown in the Heat Map chart.
Hope this helps,
Ingo
i have tried the sample markt data. not working
How to understand this result?
HOW TO CALCULATE THE AVERAGE?
tagging @IngoRM
Try to put your data into a process like this. As you can see I added an aggregation step to the automodel results so we can see some additional averages.
Also keep in mind that the features are normalized before the clustering (and hence also before the calculation of the averages).