Backward Feature Elimination End

DeprecatedKNIME Base Nodes version 3.6.1.v201808311359 by KNIME AG, Zurich, Switzerland

This is the end node of a backward feature elimination loop. It compares the target column with the prediction column set in the node's dialog. For numerical columns the squared error is computed, for all other columns the error rate. The column that caused the smallest error when left out is then discarded for the next iteration.
The node's output table shows the error for the different levels of the elimination process, and the column that has been excluded in the level.

Options

Target column

The column that contains the real values

Prediction column

The column that contains the predicted values

Input Ports

The result table of a predictor inside the loop

Output Ports

Elimination levels and corresponding error rates and left out columns

The elimination levels as model that can be used in a backward elimination filter node

Best Friends (Incoming)

• Decision Tree Predictor (14 %) Predictor
• Naive Bayes Predictor (9 %) PredictorStreamable
• Regression Predictor (9 %) PredictorDeprecated
• Random Forest Predictor (8 %) PredictorStreamableDeprecated
• Backward Feature Elimination Start (1:1) (5 %) LoopStartDeprecated
• Scorer (5 %) Other
• Weka Predictor (3.7) (5 %) Predictor
• Numeric Scorer (4 %) Other
• K Nearest Neighbor (4 %) LearnerStreamable
• MultiLayerPerceptron Predictor (4 %) PredictorStreamable
• Gradient Boosted Trees Predictor (4 %) PredictorStreamableDeprecated
• Tree Ensemble Predictor (4 %) PredictorStreamableDeprecated
• X-Aggregator (2 %) LoopEnd
• PNN Predictor (2 %) PredictorStreamable
• SVM Predictor (2 %) PredictorStreamable
• Random Forest Predictor (2 %) PredictorDeprecated
• Random Forest Predictor (Regression) (2 %) PredictorStreamable
• Backward Feature Elimination Start (2:2) (1 %) LoopStartDeprecated
• Gradient Boosted Trees Predictor (Regression) (1 %) PredictorStreamable
• Simple Regression Tree Predictor (1 %) PredictorStreamable
• Row Filter (1 %) ManipulatorStreamable
• Missing Value (1 %) Manipulator
• Backward Feature Elimination Filter (< 1 %) ManipulatorDeprecated
• Naive Bayes Learner (< 1 %) Learner
• Fuzzy Rule Predictor (< 1 %) PredictorStreamable
• Linear Regression Learner (< 1 %) Learner
• Logistic Regression Learner (< 1 %) LearnerDeprecated
• Logistic Regression Predictor (< 1 %) Predictor
• Tree Ensemble Predictor (< 1 %) PredictorStreamableDeprecated
• Random Forest Predictor (Regression) (< 1 %) PredictorDeprecated
• Missing Value Column Filter (< 1 %) Manipulator
• End IF (< 1 %) Manipulator
• K Nearest Neighbor (Distance Function) (< 1 %) Learner
• LIBSVMPredictor (< 1 %) Predictor
• Weka Predictor (3.6) (< 1 %) Predictor
• Table Reader (< 1 %) SourceStreamable
• Logistic Regression Learner (< 1 %) Learner
• SOTA Predictor (< 1 %) Predictor
• Simple Regression Tree Predictor (< 1 %) PredictorDeprecated
• Tree Ensemble Predictor (Regression) (< 1 %) PredictorStreamable
• Random Forest Learner (< 1 %) LearnerDeprecated
• String To Number (< 1 %) ManipulatorStreamable
• Column Rename (< 1 %) ManipulatorStreamable
• Rule Engine (< 1 %) PredictorStreamable
• NG Predictor (beta) (< 1 %) Predictor
• Show all 45 recommendations

Best Friends (Outgoing)

• Backward Feature Elimination Filter (78 %) ManipulatorDeprecated
• Data to Report (3 %) Other
• CSV Writer (3 %) SinkStreamable
• Model Writer (3 %) Sink
• Backward Feature Elimination Start (2:2) (3 %) LoopStartDeprecated
• Variable Loop End (3 %) LoopEnd
• Scorer (3 %) Other
• Histogram (interactive) (3 %) Visualizer
• R Snippet (3 %) Manipulator
• Logistic Regression Learner (< 1 %) Learner
• Column Filter (< 1 %) ManipulatorStreamable
• Reference Column Filter (< 1 %) ManipulatorStreamable
• Partitioning (< 1 %) Manipulator
• Line Plot (< 1 %) Visualizer
• XLS Writer (< 1 %) SinkDeprecated
• Line Plot (JavaScript) (< 1 %) Visualizer
• Show all 16 recommendations

Workflows

• 01_Feature_Elimination_with_Naive_Bayes
• 02_Techniques_for_Dimensionality_Reduction
• 002004_Feature-Elimination_with_NaiveBayes
• Dim Reduction Techniques
• KDD analysis on All Data

Update Site

To use this node in KNIME, install KNIME Base Nodes from the following update site: