backend.conformance_checking.temporal_profile

Contains functionality for temporal conformance checking.

This module defines the TemporalProfile class which uses PM4Py to discover temporal profiles from event logs and checks conformance based on the discovered temporal profiles.

Attributes

TemporalProfileType

ConformanceResultType

Classes

TemporalProfile

Represents the temporal profile of an event log.

Module Contents

type backend.conformance_checking.temporal_profile.TemporalProfileType = Dict[Tuple[str, str], Tuple[float, float]]
type backend.conformance_checking.temporal_profile.ConformanceResultType = List[List[Tuple[Any, ...]]]
class backend.conformance_checking.temporal_profile.TemporalProfile(log: pandas.DataFrame, case_id_col: str | None = None, activity_col: str | None = None, timestamp_col: str | None = None)

Represents the temporal profile of an event log.

log

The event log.

_temporal_profile

The discovered temporal profile.

_temporal_conformance_result

The result of temporal conformance checking.

_zeta

The zeta value used for temporal conformance checking.

case_id_col

The name of the Case ID column. Only needed if

Type:

optional

the log is read as a csv file.
activity_col

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

Type:

optional

timestamp_col

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

Type:

optional

log: pandas.DataFrame
case_id_col: str | None = None
activity_col: str | None = None
timestamp_col: str | None = None
discover_temporal_profile() → None

Discovers the temporal profile from the log.

The result is stored in _temporal_profile which is a dictionary where each key is a tuple of two activity names (source, target), and the value is a tuple containing:

  1. The mean duration between the two activities

  2. The standard deviation of those durations.

check_temporal_conformance(zeta: float = 0.5) → None

Checks conformance of the log against the temporal profile.

The result is stored in _temporal_conformance_result which is a list containing, for each trace, all the deviations. Each deviation is a tuple containing:

  1. The source activity of the recorded deviation.

  2. The target activity of the recorded deviation.

  3. The time passed between the occurrence of the source activity and the

    target activity.

  4. The value of (time passed - mean)/std for this occurrence (zeta).

Parameters:

zeta – Multiplier for the standard deviation.

Raises:

ValueError – If the temporal profile has not been discovered yet.

get_temporal_profile() → TemporalProfileType

Returns the discovered temporal profile.

Returns:

A dictionary where each key is a tuple of two activity names (source, target), and the value is a tuple containing:

  1. The mean duration between the two activities

  2. The standard deviation of those durations.

Raises:

ValueError – If the temporal profile has not been discovered yet.

get_temporal_conformance_result() → ConformanceResultType

Returns the result of temporal conformance checking.

Returns:

  1. The source activity of the recorded deviation.

  2. The target activity of the recorded deviation.

  3. The time passed between the occurrence of the source activity and the target activity.

  4. The value of (time passed - mean)/std for this occurrence (zeta).

Return type:

A list containing, for each trace, all the deviations. Each deviation is a tuple containing

Raises:

ValueError – If the temporal conformance result has not been computed yet.

get_zeta() → float | None

Returns the zeta value used for temporal conformance checking.

Returns:

The zeta value used for temporal conformance checking.

get_conformance_diagnostics() → pandas.DataFrame

Returns the result of temporal conformance checking as a DataFrame.

Returns:

  1. The Case ID of the recorded deviation.

  2. The source activity of the recorded deviation.

  3. The target activity of the recorded deviation.

  4. The time passed between the occurrence of the source activity and the target activity.

  5. The value of (time passed - mean)/std for this occurrence (zeta).

Return type:

A DataFrame containing the deviations for each trace. Each row contains

Raises:

ValueError – If the temporal conformance result has not been computed yet.

get_sorted_coloured_diagnostics() → pandas.io.formats.style.Styler

Returns the diagnostics DataFrame with sorting and styling.

Sorts the diagnostics DataFrame in descending order of the number of standard deviations (num_st_devs) and applies a colour-coded strip to the rows based on the value of num_st_devs - zeta:

  • Green: If the value is less than 0.5

  • Yellow: If the value is between 0.5 and 1.0

  • Red: If the value is greater than 1.0

Returns:

A styled DataFrame containing the sorted diagnostics with colour-coding.