Patent Issued for Self-training machine-learning system for generating and providing action recommendations (USPTO 11282408): UnitedHealth Group Incorporated

2022 APR 12 (NewsRx) -- By a News Reporter-Staff News Editor at Insurance Daily News -- UnitedHealth Group Incorporated (Minnetonka, Minnesota, United States) has been issued patent number 11282408, according to news reporting originating out of Alexandria, Virginia, by NewsRx editors.

The patent’s inventors are Baker, Michael (Minnetonka, MN, US), Bothra, Siddhartha (Minneapolis, MN, US), Delany, Andrew (Atlanta, GA, US), Desai, Shikha (Minneapolis, MN, US), Guthrie, Marie (Roanoke, VA, US).

This patent was filed on June 9, 2021 and was published online on March 22, 2022.

From the background information supplied by the inventors, news correspondents obtained the following quote: “In various scenarios a team may include a team leader and a plurality of team members. The team leader may be charged with coaching team members to improve the performance of the individual team members and the team as a whole. Traditionally, team leaders are tasked with identifying coaching opportunities and determining how to address such coaching opportunities. Thus, team leaders may spend a significant amount of time identifying coaching opportunities, determining how to address coaching opportunities, and addressing the coaching opportunities. This may lead to team leaders having very little time to perform other work tasks.”

Supplementing the background information on this patent, NewsRx reporters also obtained the inventors’ summary information for this patent: “BRIEF SUMMARY OF SOME EXAMPLE EMBODIMENTS

“Various embodiments provide methods, apparatuses, computer program products, systems, and/or the like that provide a team leader interactive user interface (IUI) configured to provide team leaders with a prioritized list of action items along with other information/data corresponding to team member and team performance. In various embodiments, the IUI provides team and/or team member metrics corresponding to key performance indicators (KPIs). In various embodiments, the metrics are determined by analyzing performance information/data. In various embodiments, the team and/or team member metrics are (near) real time metrics. For example, the performance information/data corresponding to the team and/or individual team members may be analyzed periodically (e.g., every hour, every two hours, and/or the like) and the metrics may be updated accordingly. In various embodiments, action items of the list of action items and/or a priority of the action items within the list of action items may be determined based on one or more of the metrics. In various embodiments, the relative priority of action items of the list of action items is used to determine the order in which the action items are presented in the list of action items. In an example embodiment, a recommendation as to address one or more action items is provided. For example, at least some of the action items may corresponding to coaching opportunities. In various embodiments, a coaching opportunity may correspond to a single defect in a team or team member’s performance (e.g., missing a single deadline), one or more comprehensive metric values (e.g., a metric corresponding to multiple tasks/interactions being below a goal level), a trend in one or more metric values, and/or the like.

“As used herein, a coaching opportunity is a situation identified where team or team member performance may be improved and/or maintained above a goal level (e.g., as measured via one or more metrics corresponding to KPIs) in response to team leader coaching. In various embodiments, team leader coaching may include any performance related communication between the team leader and one or more team members associated with the coaching opportunity. In an example embodiment, the coaching opportunity may be a positive feedback opportunity and the team leader may coach the associated team member(s) by providing positive feedback (e.g., Congratulations on making goal XYZ!, Thank you for your work on this!, etc.). In an example embodiment, the coaching opportunity may require the team leader to provide training regarding a particular procedural matter, a reminder regarding a procedural matter, a reminder of what is required to meet one or more goals, minor disciplinary action, and/or the like. In various embodiments, the coaching provided by the team leader may be based on and recorded (e.g., for various organizational records) using a coaching form. In an example embodiment, the coaching form is a form comprising information/data corresponding a coaching session. A coaching session is the interaction and/or communication through which the team leader coaches the team and/or team member(s). In an example embodiment, a coaching form comprises fields corresponding to information/data identifying the team leader, the one or more team members associated with the coaching session, one or more metrics corresponding to identifying of the coaching opportunity, feedback for provided during the coaching session (e.g., positive feedback, suggestions for improvement, procedural matter information/data, and/or the like), a date and time of the coaching session, and/or other information/data corresponding to the coaching session. In an example embodiment, a team leader may choose to forgive a coaching opportunity (e.g., choose to not address the coaching opportunity with the team and/or associated team member(s)) and/or choose to rebut a coaching opportunity (e.g., choose to argue against the coaching opportunity and/or provide additional contextual information/data corresponding to one or more metrics used to identify the coaching opportunity).

“In various embodiments, the team leader IUI provides a recommendation regarding how to address a coaching opportunity. For example, the team leader IUI may provide a coaching opportunity and a recommended strategy for responding to the coaching opportunity (e.g., provide positive feedback, address, forgive, and/or rebut). In an example embodiment, the recommendation may be determined using a recommendation model trained using machine learning. In an example embodiment, a set of teams may be clustered based on overlapping KPIs used to monitor the team’s performance and/or overlapping priorities. Completed coaching opportunities and an outcome indicator corresponding to the outcome of each of the coaching opportunities for teams of a cluster may be used to train a recommendation model for the corresponding cluster of teams. In various embodiments, a completed coaching opportunity is a coaching opportunity in which the team leader has provided positive feedback, addressed the coaching opportunity, chosen to forgive the coaching opportunity, and/or a rebuttal process has been completed. In an example embodiment, the outcome indicator may be determined within minutes of coaching opportunity being completed, one or more hours after the coaching opportunity has been completed, one or more days after the coaching opportunity has been completed, and/or the like. Through the training of the recommendation model based on coaching forms corresponding to completed coaching opportunities and the corresponding outcome indicators, the recommendation model learns to determine a recommendation for responding to various coaching opportunities that are most likely to lead to improving team and/or team member performance and/or maintaining of team and/or team member performance above a goal level. In various embodiments, the recommendation model may be further configured to determine recommendations regarding which metrics are most important for a team and/or team member to achieve one or more goals and/or to avoid one or more coaching opportunities from being initiated/identified. In various embodiments, the recommendation model may be further configured to pre-fill a coaching form to provide the team leader with information/data regarding the problem to be addressed (e.g., the one or more metrics and/or the like that triggered the coaching opportunity, trend(s) in one or more metrics that triggered the coaching opportunity, and/or the like) and how the problem may be addressed (e.g., suggestions for improving performance, training materials to be used, and/or the like).

“In various embodiments, the team leader IUI may be configured to provide a convenient dashboard through which the team leader may efficiently monitor team and/or individual team member performance (e.g., via one or more metrics corresponding to KPIs) and determine the most efficient way for improving team and/or team member performance in accordance with the priorities of the team. In particular, various embodiments provide significant improvements over the art by not only providing a team leader with a list of action items ordered based on the priorities of the team, but by also providing a recommendation regarding how the team leader may address an action item to most effectively improve the performance of the team and/or team member(s) and/or to maintain the performance of the team and/or team member(s) above a goal level. In various embodiments, the team leader IUI may provide the team leader with graphical representations of one or more metrics corresponding to KPIs for the team and/or individual team members, graphical representations of trends in one or more metrics corresponding to KPIs for the team and/or individual team members, information/data (and/or a graphical representation thereof) regarding which metrics are most important for achieving one or more team and/or team member goals, and/or the like. Thus, various embodiments aid in improving the efficiency with which team leaders may address coaching opportunities and well as improving team leader efficacy in identifying and addressing coaching opportunities.”

The claims supplied by the inventors are:

“1. A computer-implemented method comprising: providing, by one or more processors, an interactive user interface (IUI) for display via a browser executing on a user computing entity, wherein: the IUI comprises an action list of one or more action items, wherein each action item of the one or more action items corresponds to one or more team members of a team, a first action item corresponds to a first coaching opportunity and a first recommendation for responding to the first coaching opportunity, the first coaching opportunity is generated using one or more machine learning models, wherein the one or more machine learning models are trained using (a) data regarding previous handlings of coaching opportunities and corresponding outcome indicators for a cluster of teams, and (b) data regarding a plurality of teams analyzed to identify teams that have at least one of (i) overlapping sets of key performance indicator metrics used to track performance for each team or team member, or (ii) overlapping priorities to generate the cluster of teams, and the first recommendation for responding to the first coaching opportunity is generated (a) using the one or more machine learning models, and (b) based at least in part on the performance data corresponding to the plurality of key performance indicator metrics; receiving, by the one or more processors, a user selection originating from the IUI requesting a coaching form corresponding to the first coaching opportunity; providing, by the one or more processors, the coaching form for display via the IUI, wherein (a) the IUI displays the coaching form, (b) the coaching form is pre-populated with output from the one or more machine learning models, and © performance data corresponding to the plurality of key performance indicator metrics for a time period after the first team member was coached in accordance with the coaching form is used to determine one or more outcome indicators corresponding to the first coaching opportunity.

“2. The method of claim 1, wherein the coaching form and the one or more outcome indicators are used to further train the one or more machine learning models.

“3. The method of claim 1, wherein the first recommendation for responding to the first coaching opportunity is one of (a) provide positive feedback, (b) address the first coaching opportunity, or © forgive the first coaching opportunity.

“4. The method of claim 1, wherein the action list is automatically updated on a periodic basis.

“5. The method of claim 1, wherein (a) the IUI further comprises one or more metrics each corresponding to one of the plurality of key performance indicators, and (b) the one or more metrics determined by analyzing performance data corresponding to one or more team members.

“6. The method of claim 5, wherein the performance data performance data corresponding to one or more team members is analyzed in real time with respect to the generation of the performance data corresponding to one or more team members.

“7. An apparatus comprising at least one processor, at least one communications interface, a user interface, and at least one memory including computer program code, the at least one memory and computer program code configured to, with the processor, cause the apparatus to at least: provide an interactive user interface (IUI) for display via a browser executing on a user computing entity, wherein: the IUI comprises an action list of one or more action items, wherein each action item of the one or more action items corresponds to one or more team members of a team, a first action item corresponds to a first coaching opportunity and a first recommendation for responding to the first coaching opportunity, the first coaching opportunity is generated using one or more machine learning models, wherein the one or more machine learning models are trained using (a) data regarding previous handlings of coaching opportunities and corresponding outcome indicators for a cluster of teams, and (b) data regarding a plurality of teams analyzed to identify teams that have at least one of (i) overlapping sets of key performance indicator metrics used to track performance for each team or team member, or (ii) overlapping priorities to generate the cluster of teams, and the first recommendation for responding to the first coaching opportunity is generated (a) using the one or more machine learning models, and (b) based at least in part on the performance data corresponding to the plurality of key performance indicator metrics; receive a user selection originating from the IUI requesting a coaching form corresponding to the first coaching opportunity; provide the coaching form for display via the IUI, wherein (a) the IUI displays the coaching form, (b) the coaching form is pre-populated with output from the one or more machine learning models, and © performance data corresponding to the plurality of key performance indicator metrics for a time period after the first team member was coached in accordance with the coaching form is used to determine one or more outcome indicators corresponding to the first coaching opportunity.

“8. The apparatus of claim 7, wherein the coaching form and the one or more outcome indicators are used to further train the one or more machine learning models.

“9. The apparatus of claim 7, wherein the first recommendation for responding to the first coaching opportunity is one of (a) provide positive feedback, (b) address the first coaching opportunity, or © forgive the first coaching opportunity.

“10. The apparatus of claim 7, wherein the action list is automatically updated on a periodic basis.

“11. The apparatus of claim 7, wherein (a) the IUI further comprises one or more metrics each corresponding to one of the plurality of key performance indicators, and (b) the one or more metrics determined by analyzing performance data corresponding to one or more team members.

“12. The apparatus of claim 11, wherein the performance data performance data corresponding to one or more team members is analyzed in real time with respect to the generation of the performance data corresponding to one or more team members.

“13. A computer program product comprising at least one non-transitory computer-readable storage medium having computer-executable program code portions stored therein, the computer-executable program code portions, when executed by a processor of computing entity, are configured to cause the computing entity to at least: cause display of an interactive user interface (IUI) via a user interface of the computing entity, wherein: provide an IUI for display via a browser executing on a user computing entity, wherein: the IUI comprises an action list of one or more action items, wherein each action item of the one or more action items corresponds to one or more team members of a team, a first action item corresponds to a first coaching opportunity and a first recommendation for responding to the first coaching opportunity, the first coaching opportunity is generated using one or more machine learning models, wherein the one or more machine learning models are trained using (a) data regarding previous handlings of coaching opportunities and corresponding outcome indicators for a cluster of teams, and (b) data regarding a plurality of teams analyzed to identify teams that have at least one of (i) overlapping sets of key performance indicator metrics used to track performance for each team or team member, or (ii) overlapping priorities to generate the cluster of teams, and the first recommendation for responding to the first coaching opportunity is generated (a) using the one or more machine learning models, and (b) based at least in part on the performance data corresponding to the plurality of key performance indicator metrics; receive a user selection originating from the IUI requesting a coaching form corresponding to the first coaching opportunity; provide the coaching form for display via the IUI, wherein (a) the IUI displays the coaching form, (b) the coaching form is pre-populated with output from the one or more machine learning models, and © performance data corresponding to the plurality of key performance indicator metrics for a time period after the first team member was coached in accordance with the coaching form is used to determine one or more outcome indicators corresponding to the first coaching opportunity.

“14. The computer program product of claim 13, wherein the coaching form and the one or more outcome indicators are used to further train the one or more machine learning models.

“15. The computer program product of claim 13, wherein the first recommendation for responding to the first coaching opportunity is one of (a) provide positive feedback, (b) address the first coaching opportunity, or © forgive the first coaching opportunity.

“16. The computer program product of claim 13, wherein the action list is automatically updated on a periodic basis.

“17. The computer program product of claim 13, wherein (a) the IUI further comprises one or more metrics each corresponding to one of the plurality of key performance indicators, and (b) the one or more metrics determined by analyzing performance data corresponding to one or more team members.

“18. The computer program product of claim 17, wherein the performance data performance data corresponding to one or more team members is analyzed in real time with respect to the generation of the performance data corresponding to one or more team members.”

For the URL and additional information on this patent, see: Baker, Michael. Self-training machine-learning system for generating and providing action recommendations. U.S. Patent Number 11282408, filed June 9, 2021, and published online on March 22, 2022. Patent URL: http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PALL&p=1&u=%2Fnetahtml%2FPTO%2Fsrchnum.htm&r=1&f=G&l=50&s1=11282408.PN.&OS=PN/11282408RS=PN/11282408

(Our reports deliver fact-based news of research and discoveries from around the world.)