US 2021/0264332 A1 · PROCESS DISCOVERY AND OPTIMIZATION USING TIME-SERIES DATABASES, GRAPH-ANALYTICS, AND MACHINE LEARNING

Abstract

Implementations described herein relate to methods, systems, and computer-readable media to generate an alert or recommendation based on an activity log. In some implementations, a computer-implemented method includes receiving, at a processor, an activity log associated with one or more enterprise sub-processes for a first time period, determining, at the processor, change context data based on a comparison of the entity data and the associated interaction data with corresponding entity data and associated interaction data of a second time period, analyzing the change context data to determine atomic event data, providing the atomic event data associated with the enterprise process as input to a trained machine learning model, determining, using the trained machine learned model and the atomic event data, a predicted future state of the enterprise process, and generating and transmitting an alert or recommendation to one or more data sinks associated with the enterprise process.

Claims20

Dependent claims

A computer-implemented method, the method comprising:

receiving, at a processor, an activity log associated with one or more enterprise sub-processes of the enterprise process for a first time period, wherein the activity log includes entity data of one or more entities and associated interaction data of the one or more entities;
determining, at the processor, change context data based on a comparison of the entity data and the associated interaction data with corresponding entity data and associated interaction data of a second time period;
analyzing the change context data to determine atomic event data associated with the enterprise process based on detection of one or more of:
a change in state of the one or more entities or a change in state of one or more interactions of the one or more entities included in the change context data;
providing the atomic event data associated with the enterprise process as input to a trained machine learning model;
determining, using the trained machine learning model and the atomic event data, a predicted future state of the enterprise process;
generating, based on the predicted future state of the enterprise process, an alert or recommendation;
and
transmitting, the alert or recommendation to one or more data sinks associated with the enterprise process.

6 more elements

A system comprising:

a memory with instructions stored thereon;
and
a processing device, coupled to the memory, the processing device configured to access the memory and execute the instructions, wherein the instructions cause the processing device to perform operations including:
receiving, at a processor, an activity log associated with one or more enterprise sub-processes of an enterprise process for a first time period, wherein the activity log includes entity data of one or more entities and associated interaction data of the one or more entities;
determining, at the processor, change context data based on a comparison of the entity data and the associated interaction data with corresponding entity data and associated interaction data of a second time period;
analyzing the change context data to determine atomic event data associated with the enterprise process based on detection of one or more of:
a change in state of the one or more entities or a change in state of one or more interactions of the one or more entities included in the change context data;
providing the atomic event data associated with the enterprise process as input to a trained machine learning model;
determining, using the trained machine learned model and the atomic event data, a predicted future state of the enterprise process;
and
generating, based on the predicted future state of the enterprise process, an alert or recommendation.

8 more elements

Dependencies are inferred from the claim text.

Process discovery and optimization using time-series databases, graph-analytics, and machine learning

Life of this patent

Process discovery and optimization using time-series databases, graph-analytics, and machine learning

Life of this patent

Abstract

Claims20