The following node is available in the Open Source KNIME predictive analytics and data mining platform version 2.7.1. Discover over 1000 other nodes, as well as enterprise functionality at http://knime.com.

Polynomial Regression (Learner)

This node performs polynomial regression on the input data and computes the coefficients that minimize the squared error. The user must choose one column as target (dependent variable) and a number of independent variables. By default, polynomials with degree 2 are computed, which can be changed in the dialog.
If the optional PMML inport is connected and contains preprocessing operations in the TransformationDictionary those are added to the learned model.

Dialog Options

Regression settings
Target column
The column that contains the dependent "target" variable.
Polynomial degree
The maximum degree the polynomial regression function should have.
Column Selection
Select the columns containing the independent variables and move them to the "include" list.
View settings
Number of data points to show in view
This option can be use to change the number of data points in the node view if e.g. there are too many points. The default value is 10,000 points.

Ports

Input Ports
0 The input samples, which of the columns are used as independent variables can be configured in the dialog. The input must not contain missing values, you have to fix them by e.g. using the Missing Values node.
1 Optional PMML port object containing preprocessing operations.
Output Ports
0 Training data classified with the learned model and the corresponding errors.
1 The computed regression coefficients as a model for use in the Polynomial Regression Predictor.

Views

Learned Coefficients
Shows all learned coefficients all attributes.
Scatter Plot
Shows the data points and the regression function in one dimension.
This node is contained in KNIME Base Nodes provided by KNIME GmbH, Konstanz, Germany.