Statements (54)
| Predicate | Object |
|---|---|
| gptkbp:instanceOf |
Machine learning model
Supervised learning algorithm |
| gptkbp:advantage |
Can create biased trees if some classes dominate
Easy to interpret Handles both numerical and categorical data No need for feature scaling Overfitting without pruning Sensitive to small data variations |
| gptkbp:canBe |
Binary tree
Multiway tree |
| gptkbp:canBePruned |
Yes
|
| gptkbp:canBeVisualized |
Yes
|
| gptkbp:ensembleMethod |
Gradient boosting
Random forest |
| gptkbp:featureSelection |
Implicit
|
| gptkbp:hasComponent |
Branch
Internal node Leaf node Root node |
| https://www.w3.org/2000/01/rdf-schema#label |
Decision tree
|
| gptkbp:interpretable |
Yes
|
| gptkbp:introducedIn |
gptkb:Ross_Quinlan
|
| gptkbp:output |
Class label
Continuous value |
| gptkbp:popularAlgorithm |
gptkb:CHAID
gptkb:C4.5 gptkb:CART gptkb:ID3 |
| gptkbp:relatedTo |
gptkb:Bagging
Entropy Pruning Boosting Decision rules Ensemble learning Feature importance Gini index Random forest Rule-based learning |
| gptkbp:splittingCriterion |
Gini impurity
Variance reduction Gain ratio Information gain |
| gptkbp:supportsAlgorithm |
Greedy algorithm
|
| gptkbp:usedIn |
Credit scoring
Data mining Regression Predictive modeling Fraud detection Classification Customer segmentation Medical diagnosis |
| gptkbp:vulnerableTo |
Overfitting
|
| gptkbp:bfsParent |
gptkb:Bagging
|
| gptkbp:bfsLayer |
7
|