You are viewing the RapidMiner Studio documentation for version 10.1 - Check here for latest version
GLM Contribution
(Operator Toolbox)
Synopsis
This operator allows you to access the contribution of individual influence factors to your prediction.Description
Linear models are often used to not just predict but also understand the results. This operator can use the provided Generalized Linear Model and create new attributes, which define the amount of contribution of a given attribute per example.
The contribution for each attribute and for each example is calculated as: example_value_of_attribute * coefficient. The operator adds for each attribute which was used in the GL-Model a new attribute ''Contribution_$attribute_name'' with the contribution value. The resulting ExampleSet is provided at the scored output port.
Input
exa (Data Table)The example set to be evaluated.
mod (Model)The Generalized Linear Model to be used.
Output
- exa (Data Table)
The input example set with contribution attributes added.
- mod (Model)
The passed through input model.
Parameters
Tutorial Processes
Calculate contribution of marketing channels
This tutorial process demonstrate the usage of the GLM Contribution operator by applying it to a GL-Model trained on a demo marketing data set.
