Hub
  • Software
  • Blog
  • Forum
  • Events
  • Documentation
  • About KNIME
Sign in
  • KNIME Hub
  • Nodes
  • Random Forest Learner
NodeNode / Learner

Random Forest Learner

Learns a random forest*, which consists of a chosen number of decision trees. Each of the decision tree models is learned on a different set of rows (records) and a different set of columns (describing attributes), whereby the latter can also be a bit-vector or byte-vector descriptor (e.g. molecular fingerprint). The row sets for each decision tree are created by bootstrapping and have the same size as the original input table. For each node of a decision tree a new set of attributes is determined by taking a random sample of size sqrt(m) where m is the total number of attributes. 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.

Node details

Input ports
  1. Type: Table
    Input Data
    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
  1. Type: Table
    Out-of-bag Predictions
    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.
  2. Type: Table
    Attribute Statistics
    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.
  3. Type: Tree Ensembles
    Random Forest Model
    The trained model.

Extension

The Random Forest Learner node is part of this extension:

Related workflows & nodes

  1. Integrated Deployment
    machine learning education integrated deployment +1
    "Integrated Deployment" exercise for the advanced Life Science User Training - capture the data preprocessing - capture…
    knime > Education > Courses > L2-LS KNIME Analytics Platform for Data Scientists - Life Science - Advanced > Exercises > 04.1. Integrated Deployment
  2. Parameter Optimization and Cross Validation
    tree ensemble classification parameter optimization +4
    There has been no description set for this workflow's metadata.
    julian.bunzel > Public > Small examples > Parameter Optimization and Cross Validation
  3. 06_Random_Forest - Solution
    e-learning classification tree ensemble +2
    Solution to an e-learning course exercise. Train a Random Forest model to predict letters based on their image characte…
    maarit > Public > E-Learning > L2-DS KNIME Analytics Platform for Data Scientists - Advanced > Solutions > 06_Random_Forest - Solution
  4. Exercise 3_3 - Credit Scoring
    dataminer_1 > Public > UCI Applications Project > Workflows > Exercise 3_3 - Credit Scoring
  5. 06_Random_Forest - Solution
    e-learning classification tree ensemble +2
    Solution to an e-learning course exercise. Train a Random Forest model to predict letters based on their image characte…
    stervis > Public > E-Learning > L2-DS KNIME Analytics Platform for Data Scientists - Advanced > Solutions > 06_Random_Forest - Solution
  6. Exercise 2_3
    dataminer_1 > Public > UCI Applications Project > Workflows > Exercise 2_3
  7. Fraud Detection: Model Training
    fraud fraud detection random forest +4
    This workflow reads in the creditcard.csv file and trains and evaluates a Random Forest model to classify transactions …
    knime > Examples > 50_Applications > 39_Fraud_Detection > 01_Fraud_Detection_Model_Training
  8. How to use the Random Forest nodes
    This workflow shows how the random forest nodes can be used for classification and regression tasks. It also shows how …
    knime > Examples > 04_Analytics > 13_Meta_Learning > 02_Learning_a_Random_Forest
  9. Random Forest, Gradient Boosted Trees, and TreeEnsemble
    classification machine learning prediction +11
    This workflow solves a binary classification problem on the adult dataset using more advanced algorithms: - Random Fore…
    knime > Academic Alliance > Guide to Intelligent Data Science > Example Workflows > Chapter9 > 04_TreeEnsemble
  10. Training a Random Forest
    classification machine learning prediction +9
    Training a decision tree and training a random forest of decision trees.
    knime > Examples > 04_Analytics > 04_Classification_and_Predictive_Modelling > 09_Random_Forest

KNIME
Open for Innovation

KNIME AG
Hardturmstrasse 66
8005 Zurich, Switzerland
  • Software
  • Getting started
  • Documentation
  • E-Learning course
  • Solutions
  • KNIME Hub
  • KNIME Forum
  • Blog
  • Events
  • Partner
  • Developers
  • KNIME Home
  • KNIME Open Source Story
  • Careers
  • Contact us
Download KNIME Analytics Platform Read more on KNIME Server
© 2021 KNIME AG. All rights reserved.
  • Trademarks
  • Imprint
  • Privacy
  • Terms & Conditions
  • Credits