backend.conformance_checking.log_skeleton

Contains the LogSkeleton class.

This module is used to create a log skeleton from a given event log and to compute various metrics related to the log skeleton.

Classes

LogSkeleton

Represents a log skeleton.

Module Contents

class backend.conformance_checking.log_skeleton.LogSkeleton(log: pandas.DataFrame, case_id_col: str | None = None, activity_col: str | None = None, timestamp_col: str | None = None)

Represents a log skeleton.

log

The event log.

_skeleton

The log skeleton.

case_id_col

The name of the case ID column. Only needed if the log is read as csv file.

Type:

optional

activity_col

The name of the activity column. Only needed if the log is read as csv file.

Type:

optional

timestamp_col

The name of the timestamp column. Only needed if the log is read as csv file.

Type:

optional

log: pandas.DataFrame

case_id_col: str | None = None

activity_col: str | None = None

timestamp_col: str | None = None

compute_skeleton(noise_thr: float = 0.0) → None

Computes the log skeleton.

Parameters:

noise_thr – The noise threshold. Value between 0 and 1.

check_conformance_traces(traces: pandas.DataFrame) → List[Set[Any]]

Computes the conformance of traces with the log skeleton.

Parameters:

traces – A DataFrame containing the traces to be checked.

Returns:

A list of sets containing the results of the conformance. The conformance checking results for each trace include: - Outputs.IS_FIT: boolean that tells if the trace is perfectly

fit according to the model.

• Outputs.DEV_FITNESS: deviation based fitness (between 0 and 1;

  the more the trace is near to 1 the more fit is).

• Outputs.DEVIATIONS: list of deviations in the model.

get_equivalence_relation() → Set[Tuple[str, str]]

Returns the equivalence relation of the log skeleton.

Returns:

A set of tuples representing two activities that are equivalent.

get_always_after_relation() → Set[Tuple[str, str]]

Returns the always after relation of the log skeleton.

Returns:

A set of tuples representing two activities where the first activity always happens before the second activity.

get_always_before_relation() → Set[Tuple[str, str]]

Returns the always before relation of the log skeleton.

Returns:

A set of tuples representing two activities where the first activity always happens after the second activity.

get_never_together_relation() → Set[Tuple[str, str]]

Returns the never together relation of the log skeleton.

Returns:

A set of tuples representing two activities that never happen together.

get_activity_frequencies() → Dict[str, Set[int]]

Returns the activity frequencies.

For each activity, it returns the number of possible occurences per trace.

Returns:

A dictionary where the keys are the activities and the values are sets of integers representing the number of possible occurences per trace.

get_skeleton() → Dict[str, Any]

Returns the log skeleton.

Returns:

The log skeleton.