Introducing Classification Analysis in SAP Analytics Cloud Smart Predict

Objective

After completing this lesson, you will be able to Explain classification analysis in Smart Predict.

Use Cases for Classification Models

Using Augmented Analytics to Identify Potential Employee Churn

In the case study, we walk you through a scenario for using classification models to identify churn.

What sorts of topics can we investigate with a classification model?

  • Categorize credit card transactions as genuine or fraudulent.
  • Classify bank loan applications as either safe or risky.
  • Determine whether a prospect responds to a marketing campaign or not.
  • Identify employees who are likely to resign.
  • Identify customers who are at risk of canceling their mobile contracts to change provider (churn).
  • Classify if an e-mail is spam or not.
  • Determine if a machine is likely to break down in the next 60-minute interval or not.

Classification Analysis in Smart Predict

Classification analysis identifies the category that a new observation belongs to, on the basis of a training set of data containing observations whose category membership is known.

If you have data of employees who leave a role, then you could build a classification model that predicts the probability of other employees resigning.

For a classification model in SAP Analytics Cloud Smart Predict, the target variable must be binary nominal. Therefore, the target can be recorded as 1 or 0 to signify yes or no.

Three different classification analysis visualizations.

Log in to track your progress & complete quizzes