A customer has a large data set with no target variables or known results and is looking for a good approach for understanding more about groups within the data set. Which two IBM SPSS Modeler…

Question

Options

• AThe customer uses a Kohonen node in an effort to group data into clusters using a
• BThe customer uses a TwoStep node to identify the optimal set of clusters within the data.
• CThe customer uses a RFM Aggregate node to identify the optimal set of clusters within the
• DThe customer uses a Carma node in an effort to group data into clusters using a

• A
  76% (32)
• C
  7% (3)
• D
  17% (7)

Explanation

Both A and B are correct because the scenario describes unsupervised learning - no target variable, no labeled outcomes - where the goal is to discover natural groupings in the data.

• A (Kohonen) is correct because the Kohonen node implements a Self-Organizing Map (SOM), an unsupervised neural network that partitions records into clusters without requiring a target field.
• B (TwoStep) is correct because the TwoStep Cluster node automatically determines the optimal number of clusters using a two-pass algorithm, also requiring no target variable - making it ideal for exploratory grouping tasks.
• C (RFM Aggregate) is wrong because RFM (Recency, Frequency, Monetary) is a specialized marketing node designed to score customers based on transaction history, not a general-purpose clustering tool.
• D (Carma) is wrong because Carma is an association rules node (like Apriori) used to find co-occurrence patterns between items - it discovers relationships between variables, not groups of records.

Memory tip: When you see "no target variable + find groups," think clustering - Kohonen and TwoStep are SPSS Modeler's two primary clustering nodes. RFM is for retail transactions, and Carma/Apriori are for "people who buy X also buy Y" scenarios.

No community discussion yet for this question.