This node is the start of the feature selection loop. The
feature selection loop allows you to select, from all the features in
the input data set, the subset of features that is best for model
construction.
With this node you determine (i) which features/columns
are to be held
fixed in the selection process.
These constant or
"static" features/columns are included in each loop
iteration and are
exempt from elimination; (ii) which selection
strategy is to be used
on the other (variable) features/columns and
its settings; and (iii)
the specific settings of the selected strategy.
The following feature selection strategies are available:
- Forward Feature Selection is an iterative approach. It starts with having no feature selected. In each iteration, the feature that improves the model the most is added to the feature set.
- Backward Feature Elimination is an iterative approach. It starts with having all features selected. In each iteration, the feature that has on its removal the least impact on the models performance is removed.
- Genetic Algorithm is a stochastic approach that bases its optimization on the mechanics of biological evolution and genetics. Similar to natural selection, different solutions (individuals) are carried and mutated from generation to generation based on their performance (fitness). This approach converges into a local optimum and enabling early stopping might be recommended. See, e.g., this article for more insights.
- Random is a simple approach that selects feature combinations randomly. There is no converging and by chance (one of) the best feature combination will be drawn in an early iteration, so that early stopping might be recommended.
Options
Advanced Options
Popular Successors
- Partitioning23 %
- Shuffle9 %
- X-Partitioner5 %
- Linear Regression Learner5 %
- Normalizer2 %
- Random Forest Learner2 %
Linear Correlation2 %- Feature Selection Loop Start (1:1)2 %
- Column Filter2 %
- Feature Selection Filter1 %
- Naive Bayes Learner1 %
- Correlation Filter1 %
- Decision Tree Learner1 %
- Domain Calculator1 %
- Missing Value1 %
- SMOTE< 1 %
Number To String< 1 %Scorer< 1 %Gradient Boosted Trees Learner (Regression)< 1 %- Loop End< 1 %
- Random Forest Learner (Regression)< 1 %
- Random Forest Predictor< 1 %
- Column Resorter< 1 %
- Rule-based Row Filter< 1 %
- Rank Correlation< 1 %
- Statistics< 1 %
- RProp MLP Learner< 1 %
- Lag Column< 1 %
- Math Formula< 1 %
Table Row to Variable< 1 %- Naive Bayes Predictor< 1 %
- SVM Learner< 1 %
- Feature Selection Loop End< 1 %
- Logistic Regression Learner< 1 %
- Logistic Regression Predictor< 1 %
- Gradient Boosted Trees Learner (Regression)< 1 %
Random Forest Learner< 1 %- Number to String< 1 %
- One to Many< 1 %
- Missing Value Column Filter< 1 %
- H2O Local Context< 1 %
CSV Writer< 1 %- Feature Selection Loop Start (2:2)< 1 %
- Group Loop Start< 1 %
- k-Means< 1 %
- Decision Tree Predictor< 1 %
- K Nearest Neighbor< 1 %
- Polynomial Regression Learner< 1 %
- Regression Predictor< 1 %
- Scorer< 1 %
Numeric Scorer< 1 %- Gradient Boosted Trees Learner< 1 %
- Tree Ensemble Learner< 1 %
- Simple Regression Tree Learner< 1 %
- Equal Size Sampling< 1 %
Row Filter (deprecated)< 1 %- GroupBy< 1 %
- Column Splitter< 1 %
- Table to H2O< 1 %
- Column Selection Configuration< 1 %
XGBoost Tree Ensemble Learner< 1 %XGBoost Tree Ensemble Learner (Regression)< 1 %- XGBoost Predictor (Regression)< 1 %
Table Row To Variable Loop Start< 1 %Counting Loop Start< 1 %Generic Loop Start< 1 %PCA< 1 %- Reference Column Filter< 1 %
Column Rename< 1 %XGBoost Linear Ensemble Learner (Regression)< 1 %Variable to Table Row< 1 %- PNN Learner (DDA)< 1 %
- Tree Ensemble Predictor< 1 %
- Random Forest Predictor (Regression)< 1 %
- Value Counter< 1 %
- Rule-based Row Splitter< 1 %
- Linear Discriminant Analysis< 1 %
- Split Collection Column< 1 %
Table Writer< 1 %Naive Bayes Learner< 1 %- Numeric Scorer< 1 %
- Association Rule Learner< 1 %
Tree Ensemble Learner< 1 %- Bootstrap Sampling< 1 %
- Category to Number< 1 %
- Normalizer (PMML)< 1 %
- Row Sampling< 1 %
- String Replacer< 1 %
- Crosstab< 1 %
- Line Plot (legacy)< 1 %
- String Input (legacy)< 1 %
- Parameter Optimization Loop Start< 1 %
Python Script (1⇒1)< 1 %- Add Table to R< 1 %
- GET Request< 1 %
- Date&Time Difference< 1 %
- Moving Aggregator< 1 %
- Moving Average< 1 %
- RDKit Fingerprint< 1 %
- Create Collection Column< 1 %
- Extract Variables (Data)< 1 %
- Table Row to Variable< 1 %
Database Looping< 1 %- Table Writer< 1 %
File Reader< 1 %Backward Feature Elimination Start (1:1)< 1 %Loop End< 1 %- X-Aggregator< 1 %
Naive Bayes Predictor< 1 %- Cluster Assigner< 1 %
- Fuzzy c-Means< 1 %
- Hierarchical Clustering< 1 %
- DBSCAN< 1 %
K Nearest Neighbor< 1 %- MultiLayerPerceptron Predictor< 1 %
PCA Compute< 1 %Logistic Regression Learner< 1 %Logistic Regression Learner< 1 %- Entropy Scorer< 1 %
- SVM Predictor< 1 %
- PCA< 1 %
Gradient Boosted Trees Learner< 1 %- Gradient Boosted Trees Predictor (Regression)< 1 %
- Tree Ensemble Learner (Regression)< 1 %
Tree Ensemble Predictor< 1 %- Simple Regression Tree Predictor< 1 %
- Concatenate< 1 %
- Auto-Binner< 1 %
- Numeric Binner< 1 %
- Create Byte Vector< 1 %
- Cell Splitter By Position< 1 %
- Category to Number (Apply)< 1 %
- Number to Category (Apply)< 1 %
String To Number< 1 %- String to Number< 1 %
Column Appender< 1 %- Column Name Extractor< 1 %
- Constant Value Column< 1 %
- Linear Correlation< 1 %
- Table Validator< 1 %
Row Splitter (deprecated)< 1 %Joiner< 1 %- Low Variance Filter< 1 %
- Subset Matcher< 1 %
- Denormalizer< 1 %
Normalizer< 1 %- Rank< 1 %
- RowID< 1 %
- Numeric Row Splitter< 1 %
- String Manipulation< 1 %
- Target Shuffling< 1 %
- Top k Row Filter< 1 %
- Table Transposer< 1 %
- Rule Engine< 1 %
Rank Correlation< 1 %- Data Explorer< 1 %
- Numeric Outliers (Apply)< 1 %
- Numeric Outliers< 1 %
- One-way ANOVA< 1 %
- Paired t-test< 1 %
- Independent groups t-test< 1 %
- Wilcoxon-Mann-Whitney Test< 1 %
- Extract Table Dimension< 1 %
- Extract Table Spec< 1 %
- Enrichment Plotter (legacy)< 1 %
- Color Manager< 1 %
- ROC Curve (legacy)< 1 %
- Interactive Table (legacy)< 1 %
- Parallel Chunk Start< 1 %
- Similarity Search< 1 %
- Boosting Learner Loop Start< 1 %
- Boosting Predictor Loop Start< 1 %
Excel Reader (XLS)< 1 %Excel Writer (XLS)< 1 %- Dictionary Tagger< 1 %
- Weka Predictor (3.7)< 1 %
- Value Selection (legacy)< 1 %
- Table View (JavaScript)< 1 %
- Scatter Plot (JavaScript)< 1 %
- Value Selection Widget< 1 %
- Table to R< 1 %
- Data to Report (BIRT)< 1 %
- XPath< 1 %
Views
- This node has no views
Workflows
01_AnalyticsKNIME Hub- 01_AnalyticsKNIME Hub
- 01_Bike_Restocking_Alert_Training_w_Feature_SelectionKNIME Hub
- 01_Feature_Elimination_Done_ManuallyKNIME Hub
- 01_Feature_Elimination_with_Naive_BayesKNIME Hub
- Show all 141 workflows
Links
- KNIME E-Learning Course: Section 7.2. Re-executing Workflow Parts: Loopswww.knime.com/knime-introductory-course/chapter7/section2
- Feature selection-exercise- I am not understanding how to doby stelfrich on 2018-09-28forum.knime.com/p/48193
Developers
You want to see the source code for this node? Click the following button and we’ll use our super-powers to find it for you.