Machine learning techniques to predict task event

Patent metadata

Field	Value
Publication number	US-12518210-B2
Application number	US-202217847134-A
Country	US
Kind code	B2
Filing date	Jun 22, 2022
Priority date	Jun 22, 2022
Publication date	Jan 6, 2026
Grant date	Jan 6, 2026

Abstract

Machine learning techniques are disclosed for predicting a task event such as a service completion event based on a predefined workflow. In one aspect a method includes obtaining initial data for a service request (e.g., an account application), enriching the initial data with data from one or more repositories of an enterprise executing the service request, generating a data structure comprising independent variables extracted from the enriched data, receiving a request for a prediction of a completion time for the service request (e.g., an account opening event) at a first time during processing of the service request in accordance with each workflow, in response to receiving the request for the prediction, inputting the data structure into a machine-learning regression model, predicting, using the machine-learning regression model, a completion time for the service request, and providing the completion time for the service request.

First claim

What is claimed:

1 . A computer-implemented method comprising:
obtaining initial data for a service request, wherein the initial data comprises basic information captured as part of a service initiation including profile data for a user originating the service request, the service request, one or more products pertaining to the service request, and characteristics associated with the one or more products; enriching the initial data with data from one or more repositories of an enterprise executing the service request, wherein the enriching comprises augmenting the initial data with the data from the one or more repositories to generate enriched data, and wherein the data comprises a workflow for each of the one or more products, stages of each workflow, and data segments defined for the stages; generating a data structure comprising independent variables extracted from the enriched data, wherein the independent variables include a just-in-time position of the service request within each workflow; updating the data structure in real time based on a stage position change being triggered as the service request is processed through the workflow for each of the one or more products, wherein updating the data structure comprises updating the just-in-time position of the service request based on the stage position change; receiving a request for a prediction of a completion time for the service request at a first time during processing of the service request in accordance with each workflow; in response to receiving the request for the prediction, inputting the updated data structure into a machine-learning regression model; predicting, using the machine-learning regression model, a completion time for the service request, wherein the predicting comprises determining a relationship between the independent values and the completion time of the service request; and providing the completion time for the service request.

2 . The computer-implemented method of claim 1 , wherein generating the just-in-time position of the service request comprises:
upon entry of the service request into a stage of the workflow for each of the one or more products, a data entry is created in the data structure to record an ENTRY of the service request into the stage and an initial time step of 0; as the service request is processed in the stage in accordance with configured data segments and application parameters, a data entry is created in the data structure to record a REST of the service request within the stage and actual accrued duration while in processing; and when processing of the service request in the stage is completed, a data entry is created in the data structure to record an EXIT of the service request from the stage, an actual duration spent by the service request within the stage, ENTRY of the service request into a next stage, and an initial time step of 0 for the next stage.

3 . The computer-implemented method of claim 1 , further comprising receiving a change in data including accruing duration of a stay of the service request in a stage, collection of data defined by one or more of the data segments, completion of a portion of one or more data segments, or a combination thereof, and wherein the data structure is updated in accordance with the change in the data.

4 . The computer-implemented method of claim 3 , further comprising:
receiving a subsequent request for the prediction of an updated completion time for the service request at a second time during processing of the service request in accordance with each workflow; in response to receiving the subsequent request for the prediction, inputting the updated data structure into the machine-learning regression model; predicting, using the machine-learning regression model, the updated completion time for the service request, wherein the predicting comprises determining the relationship between the independent values and the updated completion time of the service request; and providing the updated completion time for the service request.

5 . The method of claim 1 , wherein the providing the completion time comprises communicating the completion time to a user.

6 . A computing system comprising:
a processor; and a memory coupled to the one or more processors, the memory storing a plurality of instructions executable by the one or more processors, the plurality of instructions comprising instructions that when executed by the one or more processors cause the one or more processors to perform processing comprising:
obtaining initial data for a service request, wherein the initial data comprises basic information captured as part of a service initiation including profile data for a user originating the service request, the service request, one or more products pertaining to the service request, and characteristics associated with the one or more products;
enriching the initial data with data from one or more repositories of an enterprise executing the service request, wherein the enriching comprises augmenting the initial data with the data from the one or more repositories to generate enriched data, and wherein the data comprises a workflow for each of the one or more products, stages of each workflow, and data segments defined for the stages;
generating a data structure comprising independent variables extracted from the enriched data, wherein the independent variables include a just-in-time position of the service request within each workflow;
updating the data structure in real time based on a stage position change being triggered as the service request is processed through the workflow for each of the one or more products, wherein updating the data structure comprises updating the just-in-time position of the service request based on the stage position change;
receiving a request for a prediction of a completion time for the service request at a first time during processing of the service request in accordance with each workflow;
in response to receiving the request for the prediction, inputting the data structure into a machine-learning regression model;
predicting, using the machine-learning regression model, a completion time for the service request, wherein the predicting comprises determining a relationship between the independent values and the completion time of the service request; and
providing the completion time for the service request.

7 . The computing system of claim 6 , wherein generating the just-in-time position of the service request comprises:
upon entry of the service request into a stage of the workflow for each of the one or more products, a data entry is created in the data structure to record an ENTRY of the service request into the stage and an initial time step of 0; as the service request is processed in the stage in accordance with configured data segments and application parameters, a data entry is created in the data structure to record a REST of the service request within the stage and actual accrued duration while in processing; and when processing of the service request in the stage is completed, a data entry is created in the data structure to record an EXIT of the service request from the stage, an actual duration spent by the service request within the stage, ENTRY of the service request into a next stage, and an initial time step of 0 for the next stage.

8 . The computing system of claim 6 , wherein the processing further comprises receiving a change in data including accruing duration of a stay of the service request in a stage, collection of data defined by one or more of the data segments, completion of a portion of one or more data segments, or a combination thereof, and wherein the data structure is updated in accordance with the c

Assignees

• Oracle Financial Services Software Ltd

Inventors

• Chauhan Shital Reprendra Singh
• Nath Mridul Kumar
• Parambath Vipesh Ambala
• Ivan Abraham
• Shree Shweta

Classifications

• G06N3/09

  Supervised learning · CPC title

• G06N20/00 Primary

  Machine learning · CPC title