US 12,277,517 B2
Automated evaluation of project acceleration
Alasdair Bailey, Perth (GB); Meghann Agarwal, Austin, TX (US); Tibor Meinczinger, Davle (CZ); and Janet Barnes, Kent (GB)
Assigned to ORACLE INTERNATIONAL CORPORATION, Redwood Shores, CA (US)
Filed by Oracle International Corporation, Redwood Shores, CA (US)
Filed on Oct. 10, 2022, as Appl. No. 17/962,677.
Application 17/962,677 is a division of application No. 16/560,414, filed on Sep. 4, 2019, granted, now 11,468,379.
Claims priority of provisional application 62/748,281, filed on Oct. 19, 2018.
Prior Publication US 2023/0034554 A1, Feb. 2, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06Q 10/06 (2023.01); G06F 3/14 (2006.01); G06N 5/02 (2023.01); G06N 7/01 (2023.01); G06N 20/00 (2019.01); G06Q 10/0631 (2023.01)
CPC G06Q 10/06313 (2013.01) [G06F 3/14 (2013.01); G06N 5/02 (2013.01); G06N 7/01 (2023.01); G06N 20/00 (2019.01)] 13 Claims
OG exemplary drawing
 
1. A computer-implemented method comprising:
storing metadata for a first subset of a plurality of projects, each having an associated value for completion and an expected end date falling within a window of time, wherein the value for completion of a given project is a quantifiable benefit that is realized when the project is completed but not when the project is incomplete;
selecting, in response to a determination that a cumulative value for completion of each project of the first subset of the plurality of projects falls below a threshold value, a second subset of the plurality of projects from a third subset of the plurality of projects, representing projects having respective expected end dates that do not fall within the window of time, wherein the selecting of the second subset of the plurality of projects from the third subset of the plurality of projects comprises:
assigning a value for completing each project in the third subset of the plurality of projects;
retrieving a set of parameters for each project in the third subset of the plurality of projects from a database implemented on a first computer system;
calculating a discount factor for the value of each project in the third subset of the plurality of projects, representing a loss of value due to completing the project on an end date within the window of time, based on the set of parameters via a client of a machine learning platform that provides access to a predictive model trained on information about a set of previous projects, stored as structured data in a repository accessible to the machine learning platform;
providing an adjusted value for each project in the third subset of the plurality of projects as a product of the assigned value for the project and the calculated discount factor;
selecting projects of the third subset of the plurality of projects, as the second subset of the plurality of projects, such that the adjusted value for completing the second subset of the plurality of projects in combination with the cumulative value for completing the first subset of the plurality of projects meets the threshold value;
changing the expected end date for each of the third subset of projects to the end date within the window of time;
determining an outcome for each of the third subset of projects; and
retraining the predictive model using the set of parameters for each of the third subset of projects and the determined outcome for each of the third subset of projects.