Learning about Predictive Analysis – Segmentation Example

Given the official announcement that BusinessObjects Predictive Analysis is now generally available, I am reviewing how I am learning about the tool following the examples from the official SAP Help (page 40) using the Retail Store Segmentation example.

Scenario as explained from SAP Help:

“The country manager of a retail chain (which has 150 stores) is finalizing plans for three sales

promotion strategies. Data pertaining to stores such as store location, sales turnover, store size, staff,

and profit margin is stored in a CSV file. The manager wants to segment 150 stores into three different

groups based on sales turnover, profit margin, store size, and staff size so that specific strategies can

be applied to each store segment.”

What is clustering, segmentation do?  I think this explanation from Microsoft explains it well:

“The algorithm uses iterative techniques to group cases in a dataset into clusters that contain similar characteristics. These groupings are useful for exploring data, identifying anomalies in the data, and creating predictions.”

First I acquire the data via a .CSV file.  Notice the integration with Visual Intelligence

While the help does not state it, I go ahead and “enrich” the attributes so turnover, size, staff, margin are now measures.

Now I select the Predict Panel (step 6 of the help) and select R-K Means

Now configure R-K means

I am configuring for 3 clusters as shown in above

Now going to the Data Writers tab (step 10 of help) I select the CSV writer to write the results of the R-K means analysis.

Then I run the Analysis and see the success message as shown below:

Deviating from the steps in the help I look at the output of the CSV file as a result of running the analysis.  Now I have a column for “Cluster”:

Here is how the results look in the Grid pane:

Now the nice part is the built-in charts offered:

In the upper left above you can see the size of the clusters in a bar chart

In the upper right you see that cluster 1 has the lowest/weakest density and cluster 3 has the highest

Lower left shows that you can select the variables and use the slider for the clusters

Lower right shows a radar chart of the clusters; it also has a slider for clusters

The above shows the algorithm summary – that is the middle icon in the Predict panel.

Additionally, you could use the visualization features from Visual Intelligence to view the results other ways.

If you are interested in learning more, there is an December 12th Webcast titled “Exciting New Release of SAP Predictive Analysis from a Leader in Analytics” (open to everyone) with SAP – register here

In January, SAP is offering a Roadmap webcast on Predictive Analysis in January 8th with Vishwanath Belur of SAP  – it is open to all with an SMP logon – register here