About scoring

In IBM® SPSS® Modeler, scoring data is defined as deploying a predictive model on new data with an unknown outcome. This predictive model processes incoming data and gives a predictive score about the likelihood or probability of an event. For example, when an online payment transaction takes place, a predictive model processes the input data and provides a predictive score which gives the probability of a transaction being either genuine or a fraud.

The normal process within SPSS Modeler is that when a predictive model receives incoming data it evaluates the input using historical data from a database and creates an output of a predicted score. This score gives a probability about an event for which a predictive analysis model is built.

The predictive model process using a scoring adapter differs from this in that the scoring adapter enables the evaluation of each record and for producing a score, or prediction, in the database without the need to export the data from database, run it through the model, and import it back again thereby making the whole process quicker.