Filters the log based on a provided set of start and end activities
The filter_endpoints
method filters cases based on the first and last activity label. It can be used in two ways: by specifying vectors with allowed start
activities and/or allowed end activities, or by specifying a percentile. In the latter case, the percentile value will be used as a cut off.
For example, when set to 0.9
, it will select the most common endpoint pairs which together cover at least 90% of the cases, and filter the log accordingly.
Usage
filter_endpoints(
log,
start_activities = NULL,
end_activities = NULL,
percentage = NULL,
reverse = FALSE,
eventlog = deprecated()
)
# S3 method for class 'log'
filter_endpoints(
log,
start_activities = NULL,
end_activities = NULL,
percentage = NULL,
reverse = FALSE,
eventlog = deprecated()
)
# S3 method for class 'grouped_log'
filter_endpoints(
log,
start_activities = NULL,
end_activities = NULL,
percentage = NULL,
reverse = FALSE,
eventlog = deprecated()
)
Arguments
- log
log
: Object of classlog
or derivatives (grouped_log
,eventlog
,activitylog
, etc.).- start_activities, end_activities
character
vector (defaultNULL
): A vector of activity identifiers, orNULL
.- percentage
numeric
(defaultNULL
): A percentage p to be used as percentile cut off. When this is used, the most common endpoint-pairs will be selected until at least the p% of the cases are selected.- reverse
logical
(defaultFALSE
): Indicating whether the selection should be reversed.- eventlog
Value
When given an object of type log
, it will return a filtered log
.
When given an object of type grouped_log
, the filter will be applied in a stratified way (i.e. each separately for each group).
The returned log will be grouped on the same variables as the original log.
Methods (by class)
filter_endpoints(log)
: Filters cases for alog
.filter_endpoints(grouped_log)
: Filters cases for agrouped_log
.
References
Swennen, M. (2018). Using Event Log Knowledge to Support Operational Exellence Techniques (Doctoral dissertation). Hasselt University.
See also
Other filters:
filter_activity()
,
filter_activity_frequency()
,
filter_activity_instance()
,
filter_activity_presence()
,
filter_case()
,
filter_case_condition()
,
filter_endpoints_condition()
,
filter_flow_time()
,
filter_idle_time()
,
filter_infrequent_flows()
,
filter_lifecycle()
,
filter_lifecycle_presence()
,
filter_precedence()
,
filter_precedence_condition()
,
filter_precedence_resource()
,
filter_processing_time()
,
filter_resource()
,
filter_resource_frequency()
,
filter_throughput_time()
,
filter_time_period()
,
filter_trace()
,
filter_trace_frequency()
,
filter_trace_length()
,
filter_trim()
,
filter_trim_lifecycle()