Using SAP Analytics Cloud Smart Discovery

Smart Discovery provides you with a powerful aid to data exploration using machine learning to explore data and uncover valuable insights. It allows you to do the following:

• Discover the key influencers driving your key performance indicators (KPIs)
• Gain insights about these influencers
• Identify outliers
• Analyze patterns in your data
• Use historical data to predict future outcomes
• Simulate what-if scenarios

Smart Discovery uses a combination of statistical analysis and predictive analytics to discover insights that would be difficult and expensive to find manually. It automatically generates a story containing the analysis of the data that can be used to drive further analysis and the visualization generated can be used in a final story.

After you run the Smart Discovery, a story is automatically created. There are 4 pages created for each story:

1. Overview
2. Key Influencers
3. Unexpected values
4. Simulation

The Key Influencers page in the story shows the most important factors influencing your target variable, in this example, the TOTAL_SALARY. It is possible to select the bars in the chart to see which other variables are correlated influencers. You can run Smart Insights on the bars of the chart to further explore the data.

The Unexpected Values report compares the actual to the expected values. In the example, you can see that there are unexpected values listed. The difference column indicates which employees are paid more or less than the expected rate which can be used to build an analysis of any salary disparities.

In the simulation report, you can modify the value of the influencers to analyze how they change the target value. It is a way to perform what if scenarios using the data and immediately see the impact on the target value of the report.