Learning about Automated Clustering with SAP Predictive Analytics 2.0

I watched this video to get started with SAP Predictive Analytics:

I was interested in the automated clustering; you can do clustering analysis with the Expert Analytics part of Predictive but this time, I thought, why not let the tool “do it for me”?

Figure 1

First (and this took the longest time) when I selected type “Excel” it is looking for an .XLS file, not XLSX

So I found a sample file that was .XLS

Figure 2

Figure 2 shows you can upload a description file (field names and lengths) – I just went with “Analyze” and fortunately it read the headings and length of the first row of the Excel file.

Figure 3

Figure 3 shows the results of Analyzing

Figure 4

I didn’t select any target variables or weighted variables.  I select 5 (minimum) to 10 (maximum) clusters

Figure 5

The above shows the results and decides that 5 clusters is “winner”.

Figure 6

Figure 6 shows the cluster frequencies

Figure 7

Figure 7 shows another cluster frequencies report that is available

Figure 8

Figure 8 is a bubble chart offered.  What could you do with this?  Possibly if you are a US tax service, you could offer your tax counseling advice to those at age 42 with high capital gains.

Once I got past the confusion of analyzing the descriptions and to use an old version of Excel, this tool was very simple to use; simply load a file and start analyzing.  You can check the other predictive analytics tutorials here Official Product Tutorials – SAP Predictive Analytics, SAP Predictive Analysis and SAP InfiniteInsight