IBM SPSS Modeler 18.2 Algorithms Guide helps organizations improve customer and citizen relationships through an in-depth understanding of data.
This guide documents algorithms used by IBM SPSS Modeler 18.2, an enterprise data mining workbench. It describes how adjusted propensity scores are calculated for different model types.
It also explains the anomaly detection algorithm, including its modeling, scoring, and reasoning stages for identifying unusual cases.
Depending on the model type, either a model-dependent method or a general-purpose method is used.
It is designed to quickly detect unusual cases for data-auditing purposes during exploratory data analysis.
The neural network uses one hidden layer with 3 neurons and persistence set to 200.