pm4py.algo.discovery.causal.variants.alpha module
This module contains code that allows us to compute a causal graph, according to the alpha miner.
It expects a dictionary of the form (activity,activity) -> num of occ.
A causal relation holds between activity a and b, written as a->b, if dfg(a,b) > 0 and dfg(b,a) = 0.
-
pm4py.algo.discovery.causal.variants.alpha.apply(dfg: Dict[Tuple[str, str], int]) → Dict[Tuple[str, str], int][source]
Computes a causal graph based on a directly follows graph according to the alpha miner
- Parameters:
dfg (dict directly follows relation, should be a dict of the form (activity,activity) -> num of occ.)
- Returns:
causal_relation
- Return type:
dict containing all causal relations as keys (with value 1 indicating that it holds)
