A One Rule or decision stump is a decision tree that consists of a single fork node with two outcomes. When run against a model that includes a number of predictor variables, the One Rule fork node simply chooses the single predictor variable that most accurately enables the choice between the two alternatives and ignores everything else. Like Zero Rule, One Rule is useful mostly as a basemark for more complex classification algorithms, which must necessarily outperform it to be of any use.
- alias
- Decision Stump
- subtype
- has functional building block
- FBB_Classification
- has input data type
- IDT_Vector of categorical variables IDT_Vector of quantitative variables
- has internal model
- INM_Rule
- has output data type
- ODT_Classification
- has learning style
- LST_Supervised
- has parametricity
- PRM_Parametric
- has relevance
- REL_Benchmark
- uses
- sometimes supports
- mathematically similar to