Learns a random forest*, which consists of a chosen number of decision trees. Each of the decision tree models is built with a different set of rows (records) and for each split within a tree a randomly chosen set of columns (describing attributes) is used. The row sets for each decision tree are created by bootstrapping and have the same size as the original input table. The attribute set for an individual split in a decision tree is determined by randomly selecting sqrt(m) attributes from the available attributes where m is the total number of learning columns. The attributes can also be provided as bit (fingerprint), byte, or double vector. The output model describes a random forest and is applied in the corresponding predictor node.
This node provides a subset of the functionality of the Tree Ensemble Learner corresponding to a random forest. If you need additional functionality please check out the Tree Ensemble Learner.
Experiments have shown the results on different datasets are very similar to the random forest implementation available in R.
The decision tree construction takes place in main memory (all data and all models are kept in memory).
The missing value handling corresponds to the method described here. The basic idea is that for each split to try to send the missing values in every possible direction; the one yielding the best results (i.e. largest gain) is then used. If no missing values are present during training, the direction of the split that the most records are following is chosen as the direction for missing values during testing.
Nominal columns are split in a binary manner. The determination of the split depends on the kind of problem:
- For two-class classification problems the method described in section 9.4 of "Classification and Regression Trees" by Breiman et al. (1984) is used.
- For multi-class classification problems the method described in "Partitioning Nominal Attributes in Decision Trees" by Coppersmith et al. (1999) is used.
(*) RANDOM FORESTS is a registered trademark of Minitab, LLC and is used with Minitab’s permission.
Options
Input Ports
- The data to be learned from. They must contain at least one nominal target column and either a fingerprint (bit-vector/byte-vector) column or another numeric or nominal column.
Output Ports
- The input data with the out-of-bag predictions, i.e. for each input row this is the majority vote of all models that did not use the row during their training. The appended columns are equivalent to the columns appended by the corresponding predictor node. There is one additional column model count, which contains the number of models used for the voting (number of models not using the row throughout learning.) The out-of-bag predictions can be used to get an estimate of the generalization error of the random forest by feeding them into the Scorer node.
- A statistics table on the attributes used in the different trees. Each row represents one training attribute with these statistics: #splits (level x) as the number of models, which use the attribute as split on level x (with level 0 as root split); #candidates (level x) is the number of times an attribute was in the attribute sample for level x (in a random forest setup these samples differ from node to node). If no attribute sampling is used #candidates is the number of models. Note, these numbers are uncorrected, i.e. if an attribute is selected on level 0 but is also in the candidate set of level 1 (but will not be split on level 1 because it has been split one level up), the #candidate number will still count the attribute as candidate.
- The trained model.
Popular Predecessors
- Partitioning18 %
- Parameter Optimization Loop Start5 %
- X-Partitioner5 %
- Domain Calculator3 %
- SMOTE3 %
- Column Filter3 %
- Normalizer2 %
- Number to String2 %
- Missing Value2 %
- Equal Size Sampling2 %
- Scorer2 %
- CSV Reader1 %
Table Row to Variable1 %Column Appender1 %- Feature Selection Loop Start (1:1)< 1 %
- Category to Number (Apply)< 1 %
- String to Number< 1 %
Table Reader< 1 %- Fuzzy c-Means< 1 %
- Table Row to Variable< 1 %
- Excel Reader< 1 %
String To Number< 1 %Column Rename< 1 %- Feature Selection Filter< 1 %
- Concatenate< 1 %
- Rule Engine< 1 %
- Bootstrap Sampling< 1 %
Scorer< 1 %- Nominal Value Row Splitter< 1 %
- Molecule Type Cast< 1 %
Counting Loop Start< 1 %- Low Variance Filter< 1 %
Table Row To Variable Loop Start< 1 %- K Nearest Neighbor< 1 %
- Random Forest Learner (Regression)< 1 %
- Correlation Filter< 1 %
- Normalizer (PMML)< 1 %
Merge Variables< 1 %- Math Formula (Variable)< 1 %
- Target Shuffling< 1 %
- Column Filter (legacy)< 1 %
- File Reader< 1 %
- Table Row to Variable Loop Start< 1 %
- Python Script (legacy)< 1 %
- Random Forest Predictor< 1 %
Number To String (PMML)< 1 %- Color Manager< 1 %
- Decision Tree Predictor< 1 %
- Gradient Boosted Trees Predictor< 1 %
- Random Forest Learner< 1 %
- Number Rounder< 1 %
- RDKit Fingerprint< 1 %
- One to Many (PMML)< 1 %
CSV Reader< 1 %- Row Sampling< 1 %
- ROC Curve (JavaScript)< 1 %
- One to Many< 1 %
Linear Correlation< 1 %- Duplicate Row Filter< 1 %
- Decision Tree View (JavaScript)< 1 %
- Joiner< 1 %
File Reader< 1 %- Chunk Loop Start< 1 %
- Double to Integer< 1 %
Row Splitter (deprecated)< 1 %- Data Generator< 1 %
Python Script (1⇒1)< 1 %- RDKit Descriptor Calculation< 1 %
- Group Loop Start< 1 %
- Cluster Assigner< 1 %
- Tree Ensemble Learner< 1 %
- Column Splitter< 1 %
- String Replacer< 1 %
- Table Column to Variable< 1 %
- Parameter Optimization Loop Start (Table)< 1 %
- Merge Variables< 1 %
- Auto-Binner< 1 %
- Numeric Binner< 1 %
- Edit Nominal Domain< 1 %
Row Filter (deprecated)< 1 %- String Manipulation< 1 %
- Statistics< 1 %
- ARFF Reader< 1 %
- Feature Selection Loop Start (2:2)< 1 %
- Naive Bayes Learner< 1 %
- Numeric Scorer< 1 %
- PCA< 1 %
- Constant Value Column Filter< 1 %
- Missing Value Column Filter< 1 %
- Nominal Value Row Filter< 1 %
- Denormalizer< 1 %
- Number to String (PMML)< 1 %
- Shuffle< 1 %
- Rule-based Row Filter< 1 %
- Rule-based Row Splitter< 1 %
- Rank Correlation< 1 %
- Capture Workflow Start< 1 %
- Text Classifier Model to Table< 1 %
- X-Aggregator< 1 %
- Logistic Regression Learner< 1 %
- Category to Number< 1 %
- Column Resorter< 1 %
- GroupBy< 1 %
- Normalizer (Apply)< 1 %
- Value Counter< 1 %
- Rule Engine Variable< 1 %
- ROC Curve (legacy)< 1 %
- Math Formula< 1 %
- Document Data Extractor< 1 %
- Table View (JavaScript)< 1 %
- R Snippet< 1 %
Table Column to Variable< 1 %Table Row to Variable< 1 %- Table Creator< 1 %
- k-Means< 1 %
- SVM Predictor< 1 %
- PCA Apply< 1 %
Cell Replacer< 1 %- Cell Splitter< 1 %
- Column Combiner< 1 %
- Number to Category (Apply)< 1 %
- Lag Column< 1 %
- Column Merger< 1 %
- Linear Correlation< 1 %
- Reference Column Filter< 1 %
- Reference Row Filter< 1 %
- Reference Row Splitter< 1 %
- Missing Value (Apply)< 1 %
- Decision Tree to Ruleset< 1 %
- Data Explorer< 1 %
- Numeric Outliers< 1 %
- DB Reader< 1 %
- Model Loop Start< 1 %
- Boosting Learner Loop Start< 1 %
- Document Vector< 1 %
- Scorer (JavaScript)< 1 %
- Parameter Optimization Loop End< 1 %
- Conformal Calibration Loop Start< 1 %
- Table Reader< 1 %
- Fuzzy Rule Learner< 1 %
- Linear Regression Learner< 1 %
- Column Appender< 1 %
- Table Manipulator< 1 %
- Column Expressions< 1 %
- Strings to Document< 1 %
- Loop End< 1 %
- XGBoost Tree Ensemble Learner< 1 %
- Counting Loop Start< 1 %
Gradient Boosted Trees Predictor< 1 %Joiner< 1 %- Integer Input (legacy)< 1 %
- Outlier Removal< 1 %
Table Writer< 1 %- Naive Bayes Predictor< 1 %
- Logistic Regression Predictor< 1 %
- Column Auto Type Cast< 1 %
- Constant Value Column< 1 %
- Edit Numeric Domain< 1 %
- Column Filter (PMML)< 1 %
CASE Switch Data (End)< 1 %Color Appender< 1 %- Ungroup< 1 %
- MoSS< 1 %
- Java Snippet< 1 %
PCA Apply< 1 %- Auto-Binner (Apply)< 1 %
- RowID< 1 %
Excel Reader (XLS)< 1 %Python Script (1⇒2)< 1 %- MolSimilarity< 1 %
- Strings to Binary Objects< 1 %
Loop End< 1 %- Column Rename (Regex)< 1 %
String To Number (PMML)< 1 %Rank Correlation< 1 %- DB Query Reader< 1 %
- DB Writer< 1 %
- Category to Class< 1 %
- Integer Configuration< 1 %
- Value Selection Widget< 1 %
- Fingerprints< 1 %
- CASE Switch Start< 1 %
- K Nearest Neighbor (Distance Function)< 1 %
- Date&Time-based Row Filter< 1 %
- Split Collection Column< 1 %
Variable to Table Row< 1 %- Hierarchical Clustering< 1 %
Concatenate (Optional in)< 1 %- Sorter< 1 %
Flow Variable End IF< 1 %- Read Variables< 1 %
- Fingerprints Expander< 1 %
Active Learn Loop Start< 1 %- Create Collection Column< 1 %
- Binary Objects to Strings< 1 %
- Column List Loop Start< 1 %
Naive Bayes Learner< 1 %Numeric Scorer< 1 %- Tree Ensemble Learner (Regression)< 1 %
- Random Forest Predictor (Regression)< 1 %
- Expand Bit Vector< 1 %
- Pivot< 1 %
- Ruleset Editor< 1 %
- End IF< 1 %
CASE Switch Data (Start)< 1 %- Conditional Label Assigner< 1 %
- Random Item Inserter< 1 %
- Variable Expressions< 1 %
Excel Reader (XLS)< 1 %- Bag Of Words Creator< 1 %
- Document Vector Hashing< 1 %
- Weka Predictor (3.7)< 1 %
- Column Filter Configuration< 1 %
- Image to DataRow< 1 %
- Shapley Values Loop Start< 1 %
- Box Plot (legacy)< 1 %
- SDF Reader< 1 %
- Double Configuration< 1 %
- Date&Time Shift< 1 %
- Twitter Users< 1 %
- Java Edit Variable< 1 %
Number To String< 1 %- IF Switch< 1 %
- String to Date&Time< 1 %
- RDKit Count-Based Fingerprint< 1 %
- Fast Fourier Transform (FFT)< 1 %
- Molecule Properties< 1 %
- Number Formatter< 1 %
- Normalize Plates (NPI)< 1 %
- URI Port to Variable< 1 %
- Wait...< 1 %
Table Column to Variable< 1 %Database Connection Table Reader< 1 %Database Reader< 1 %Interval Loop Start< 1 %Generic Loop Start< 1 %Recursive Loop Start (2 ports)< 1 %- PNN Predictor< 1 %
- Decision Tree Learner< 1 %
- MDS< 1 %
- MultiLayerPerceptron Predictor< 1 %
- RProp MLP Learner< 1 %
PCA< 1 %- Regression Predictor< 1 %
- PCA Compute< 1 %
- Random Forest Distance< 1 %
PMML Gradient Boosted Trees Predictor< 1 %- Tree Ensemble Predictor< 1 %
- Add Empty Rows< 1 %
- Binner (Dictionary)< 1 %
- Create Bit Vector< 1 %
- Column Comparator< 1 %
- Cross Joiner< 1 %
- Normalizer Apply (PMML)< 1 %
- String to Number (PMML)< 1 %
- Numeric Row Splitter< 1 %
- String Manipulation (Variable)< 1 %
String Replace (Dictionary)< 1 %- Table Transposer< 1 %
- Numeric Outliers (Apply)< 1 %
- One-way ANOVA< 1 %
CASE Switch Variable (Start)< 1 %- Cache< 1 %
- Spark to Table< 1 %
- Fingerprint Bayesian Learner< 1 %
- Parallel Chunk Start< 1 %
- Gamma Distributed Assigner< 1 %
- PMML Predictor< 1 %
- Prediction Fusion< 1 %
- Math Formula (Multi Column)< 1 %
- Molecule String Input (legacy)< 1 %
Read Excel Sheet Names (XLS)< 1 %- Java Edit Variable (simple)< 1 %
- Word Vector Apply (legacy)< 1 %
- Topic Extractor (Parallel LDA)< 1 %
- Snowball Stemmer< 1 %
- RSS Feed Reader< 1 %
- Boolean Configuration< 1 %
- Value Selection Configuration< 1 %
- Column Selection (legacy)< 1 %
- Scatter Plot (JavaScript)< 1 %
- Integer Widget< 1 %
- String Widget< 1 %
Container Input (Variable)< 1 %- JSON to Table< 1 %
- Image Normalizer< 1 %
- Connected Component Analysis< 1 %
- Find Edges< 1 %
- Python Source (legacy)< 1 %
- GET Request< 1 %
Extract Date&Time Fields< 1 %- Moving Average< 1 %
- XGBoost Predictor< 1 %
- Molecular Properties< 1 %
- Run Garbage Collector< 1 %
- Run Heavy Garbage Collector< 1 %
- Maccs Fingerprint< 1 %
- Extract Variables (Data)< 1 %
- Variable to Table Row< 1 %
- Generic Loop Start< 1 %
- PNN Learner (DDA)< 1 %
- DBSCAN< 1 %
- Association Rule Learner< 1 %
- Tree Ensemble Model Extract< 1 %
- Create Byte Vector< 1 %
- Edit Nominal Domain (Dictionary)< 1 %
- Filter Apply Row Splitter< 1 %
- Many to One (PMML)< 1 %
- Single sample t-test< 1 %
- Crosstab< 1 %
- Capture Workflow End< 1 %
- TimeDelay Loop Start< 1 %
- Boosting Predictor Loop Start< 1 %
- Term to String< 1 %
- Nominal Row Filter Configuration< 1 %
- Credentials Configuration< 1 %
- String Configuration< 1 %
- Value Filter (legacy)< 1 %
- Double Input (legacy)< 1 %
- Double Widget< 1 %
- Column Selection Widget< 1 %
- JSON to XML< 1 %
- R Source (Table)< 1 %
- Date&Time Part Extractor< 1 %
- Remove Empty Rows< 1 %
Views
- Tree Views
- An decision tree viewer for all the trained models. Use the spinner to iterate through the different models.
Workflows
!Modern UI_ABC-XYX-RFM AutoML SuperstoreKNIME Hub- [Latest] Consolidated 3 Mar 1210KNIME Hub
- 0_install_AI_extensionsKNIME Hub
- 01 Advanced Machine Learning - ExerciseKNIME Hub
- 01 Advanced Machine Learning - solutionKNIME Hub
- Show all 831 workflows
Links
- No links available
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.