US 12,105,725 B2
Automatic determination of alternative paths for a process flow using machine learning
Zineb Laraki, Atherton, CA (US); Penny Tselikis, San Francisco, CA (US); Till Christian Bergmann, San Mateo, CA (US); Michael Weil, New York, NY (US); Christian Posse, Belmont, CA (US); Jason Teller, Arcadia, CA (US); Alex Edelstein, San Francisco, CA (US); Sarah Joann Aerni, San Francisco, CA (US); and Mehmet Ezbiderli, San Francisco, CA (US)
Assigned to Salesforce, Inc., San Francisco, CA (US)
Filed by Salesforce, Inc., San Francisco, CA (US)
Filed on Jan. 31, 2022, as Appl. No. 17/589,778.
Prior Publication US 2023/0244686 A1, Aug. 3, 2023
Int. Cl. G06F 16/25 (2019.01); G06F 16/23 (2019.01); G06F 16/2457 (2019.01); G06N 20/00 (2019.01)
CPC G06F 16/254 (2019.01) [G06F 16/23 (2019.01); G06F 16/2457 (2019.01); G06N 20/00 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A method for automatic process mining at one or more servers, comprising:
filtering out, for a process mining procedure, a subset of log entries corresponding to detailed actions from an event log comprising a set of log entries for a tenant of a multi-tenant database system to obtain a filtered event log;
identifying, from the filtered event log, a pattern of log entries corresponding to main actions, the pattern of log entries comprising a plurality of first log entries performed in a specific order, wherein the pattern of log entries occurs, in the filtered event log, a quantity of times satisfying a frequency threshold for the process mining procedure;
identifying, from the event log, a plurality of log entries associated with the identified pattern of log entries, the plurality of log entries comprising the plurality of first log entries in the identified pattern of log entries corresponding to the main actions and one or more second log entries corresponding to the detailed actions, the one or more second log entries associated with at least one of the plurality of first log entries in the identified pattern of log entries, wherein a main action comprises an action selected by a user for the multi-tenant database system to perform, and a detailed action comprises an automated action performed by the multi-tenant database system;
retrieving data from the multi-tenant database system corresponding to a history field of a data object associated with the plurality of log entries;
determining at least a portion of a process flow for the data object according to the identified pattern of log entries and based at least in part on the plurality of log entries and the data corresponding to the history field, the process flow comprising one or more operations to perform using the data object; and
transmitting, to a user device, an indication of at least the portion of the process flow.