Cognitive prioritization for report generation

Patent metadata

Field	Value
Publication number	US-11295257-B2
Application number	US-201815954754-A
Country	US
Kind code	B2
Filing date	Apr 17, 2018
Priority date	Apr 17, 2018
Publication date	Apr 5, 2022
Grant date	Apr 5, 2022

Abstract

A system for cognitive prioritization for report generation may include a processor and a memory cooperating therewith. The processor may be configured to accept a request for a new report from a user, the request having a user profile importance associated therewith and generate a predicted completion time for the new report based upon a historical completion time prediction model based upon historical data for prior reports. The processor may be configured to generate a predicted importance of the new report based upon a historical importance prediction model based upon the historical data for prior reports and determine a combined predicted importance based upon the user profile importance and the predicted importance. The processor may also be configured to generate a prioritization of the new report among other reports based upon the predicted completion time and the combined predicted importance and generate the new report based upon the prioritization.

First claim

What is claimed is:

1. A system for cognitive prioritization for report generation comprising:
a processor and a memory cooperating therewith, the processor configured to:
accept a request for a new report from a user, the request having a user profile importance associated therewith and a user comment indicating a reason for requesting the new report;
generate a predicted completion time for the new report by a historical completion time prediction model trained using historical data for prior reports;
generate a predicted importance of the new report based upon the reason for requesting the new report by a historical importance prediction model including a text mining model trained using the historical data for prior reports, wherein the historical data for prior reports includes user comments indicating reasons for requesting the prior reports and corresponding labeled importance;
determine a combined predicted importance based upon the user profile importance and the predicted importance;
determine that the user misstated importance of the new report in the reason for requesting the new report and adjust the predicted completion time for the new report to lower a prioritization of the new report by:
determining a number of attempts of the user to misstate importance of the new report and the prior reports in the reason for requesting the new report in the user comment for the new report and in the reasons for requesting the prior reports in the user comments for the prior reports;
determining a delay based on the number of attempts of the user to misstate importance of the new report and the prior reports, wherein the delay increases as the number of attempts increases; and
adding the delay to the predicted completion time to indicate a later time for completion of the new report;

generate the prioritization of the new report among a plurality of other reports based upon the predicted completion time and the combined predicted importance; and
control wait times for users for the report generation according to report urgency, wherein controlling the wait times includes:
using differing amounts of computing and memory resources of the processor and the memory for the report generation in a resource constrained environment having computing and memory resource capacities, wherein the amount of computing and memory resources used is according to an order of the prioritization of the new report and the plurality of other reports; and
generating the new report and the plurality of other reports in the resource constrained environment in the order according to the prioritization using the differing amounts of the computing and memory resources.

2. The system of claim 1 wherein the historical data for prior reports comprises metadata of prior reports.

3. The system of claim 1 wherein the new report has metadata associated therewith.

4. The system of claim 1 wherein the historical completion time prediction model comprises a multi-variable regression model trained on historical reports each having associated completion times.

5. The system of claim 4 wherein attributes of the multi-variable regression model comprise user-chosen values for each of a plurality of usable filters for generation of the new report, available system resources for the new report generation, and demand queue size at a time of the request for the new report.

6. The system of claim 1 wherein the user profile importance comprises a user's organization role as an input or is learned.

7. The system of claim 1 wherein the processor is configured to determine the combined predicted importance based upon the user profile importance and the predicted importance based upon one of an ensemble function and an aggregation function.

8. The system of claim 1 wherein the processor is configured to generate the prioritization of the new report by calculating an importance weight for the new report by combining a reciprocal of the predicted completion time for the new report and the predicted importance.

9. A method for cognitive prioritization for report generation comprising:
using a processor cooperating with a memory to:
accept a request for a new report from a user, the request having a user profile importance associated therewith and a user comment indicating a reason for requesting the new report;
generate a predicted completion time for the new report by a historical completion time prediction model trained using historical data for prior reports;
generate a predicted importance of the new report based upon the reason for requesting the new report by a historical importance prediction model including a text mining model trained using the historical data for prior reports, wherein the historical data for prior reports includes user comments indicating reasons for requesting the prior reports and corresponding labeled importance;
determine a combined predicted importance based upon the user profile importance and the predicted importance;
determine that the user misstated importance of the new report in the reason for requesting the new report and adjust the predicted completion time for the new report to lower a prioritization of the new report by:
determining a number of attempts of the user to misstate importance of the new report and the prior reports in the reason for requesting the new report in the user comment for the new report and in the reasons for requesting the prior reports in the user comments for the prior reports;
determining a delay based on the number of attempts of the user to misstate importance of the new report and the prior reports, wherein the delay increases as the number of attempts increases; and
adding the delay to the predicted completion time to indicate a later time for completion of the new report;

generate the prioritization of the new report among a plurality of other reports based upon the predicted completion time and the combined predicted importance; and
control wait times for users for the report generation according to report urgency, wherein controlling the wait times includes:
using differing amounts of computing and memory resources of the processor and the memory for the report generation in a resource constrained environment having computing and memory resource capacities, wherein the amount of computing and memory resources used is according to an order of the prioritization of the new report and the plurality of other reports; and
generating the new report and the plurality of other reports in the resource constrained environment in the order according to the prioritization using the differing amounts of the computing and memory resources.

10. The method of claim 9 wherein the historical completion time prediction model comprises a multi-variable regression model trained on historical reports each having associated completion times.

11. The method of claim 10 wherein attributes of the multi-variable regression model comprise user-chosen values for each of a plurality of usable filters for generation of the new report, available system resources for the new report generation, and demand queue size at a time of the request for the new report.

12. The method of claim 9 wherein using the processor comprises using the processor to determine the combined predicted importance based upon the user profile importance and the predicted importance based upon one of an ensemble function and an aggregation function.

13. The method of claim 9 wherein using the processor comprises using the processor to generate the prioritization of the new report by calculating an importance weight for the new report by combining a reciprocal of

Assignees

• IBM

Inventors

• Asthana Shubhi
• Becker Valeria
• Gajananan Kugamoorthy
• Megahed Aly

Classifications

• G06Q10/06316 Primary

  Sequencing of tasks or work · CPC title

• G06Q10/10

  Office automation; Time management · CPC title

• G06N20/20

  Ensemble learning · CPC title

• G06N20/00

  Machine learning · CPC title

• G06N5/04

  Inference or reasoning models · CPC title