pairwise_transitivity
mcda.pairwise_transitivity(report)
Test the pairwise majority relation for one consistent order.
Reads a PairwiseSuperiorityReport and builds the pairwise majority relation from its per-pair outperformance counts: method i outperforms j when it does so on more of their shared datasets than j does. Those counts already apply the region of practical equivalence, so a difference inside the noise floor adds no edge. The relation is then checked for a method preferred to all others, for circular triads, and for transitivity, and the coefficient of consistence is reported when every pair is decided.
Parameters
| Name | Type | Description | Default |
|---|---|---|---|
report |
PairwiseSuperiorityReport | A PairwiseSuperiorityReport from beam.mcda.pairwise_superiority. |
required |
Returns
| Type | Description |
|---|---|
| PairwiseTransitivityReport |
Examples
>>> import numpy as np
>>> from beam.mcda import pairwise_superiority, pairwise_transitivity
>>> scores = np.array([[0.9, 0.8, 0.7], [0.5, 0.6, 0.4], [0.2, 0.1, 0.3]])
>>> sup = pairwise_superiority(scores, "higher_is_better")
>>> cc = pairwise_transitivity(sup)
>>> cc.condorcet_choice
0
>>> cc.is_transitive
True
>>> cc.n_circular_triads
0