Quantcast
Channel: SCN : All Content - SAP BusinessObjects Predictive Analytics
Viewing all articles
Browse latest Browse all 836

Conclusions with Confusion - or Decision Trees

$
0
0

Following up from Learning about Clustering Analysis with Predictive Analysis it looks like cluster 3 is the most valuable.  To validate, I look at Decision Trees in Predictive Analysis 1.14:

 

 

1setup.png

 

I set it up as trend, classification and the 4 key measures of rental price, count, revenue and margin.  I also read up about decision trees in Wikipedia

 

The target variable is the cluster number from the R K means analysis.  After running the analysis one of the visualizations available is the confusion matrix:

 

 

2confusionmatrix.png

 

Of course to learn more about this I turn to Wikipedia to learn more about Confusion Matrix

 

 

 

4couldcluster2 be more valuable.png

The decision tree above shows that cluster 2 may be more valuable than cluster 3, so maybe I should target those customers.


Viewing all articles
Browse latest Browse all 836

Trending Articles



<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>