Docs

Reclassifying the data

Last updated: Oct 09, 2024
Reclassifying the data (SPSS Modeler)
Figure 1. Example flow showing string reclassification for binomial logistic regression
Example flow showing string reclassification for binomial logistic regression
  1. Add a Data Asset node that points to drug_long_name.csv.
  2. Add a Type node after the Data Asset node. Double-click the Type node to open its properties, and select Cholesterol_long as the target.
  3. Add a Logistic Regression node after the Type node. Double-click the node and select the Binomial procedure (instead of the default Multinomial procedure).
  4. Right-click the Logistic Regression node and run it. An error message warns you that the Cholesterol_long string values are too long. When you encounter this type of message, follow the procedure described in the rest of this example to modify your data.
    Figure 2. Error message displayed when running the binomial logistic regression node
    Error message displayed when running the binomial logistic regression node
  5. Add a Reclassify node after the Type node and double-click it to open its properties.
  6. For the Reclassify Field, select Cholesterol_long and type Cholesterol for the new field name.
  7. Click Get values to add the Cholesterol_long values to the original value column.
  8. In the new value column, type High next to the original value of High level of cholesterol and Normal next to the original value of Normal level of cholesterol.
    Figure 3. Reclassifying long strings
    Reclassifying long strings
  9. Add a Filter node after the Reclassify node. Double-click the node, choose Filter the selected fields, and select the Cholesterol_long field.
    Figure 4. Filtering the "Cholesterol_long" field from the data
    Filtering the "Cholesterol_long" field from the data
  10. Add a Type node after the Filter node. Double-click the node and select Cholesterol as the target.
    Figure 5. Short string details in the "Cholesterol" field
    Short string details in the "Cholesterol" field
  11. Add a Logistic node after the Type node. Double-click the node and select the Binomial procedure.

You can now run the binomial Logistic node and generate a model without encountering the error as you did before.

This example only shows part of a flow. For more information about the types of flows in which you might need to reclassify long strings, see the following example: