US 12,243,638 B1
Generating service offerings based on associated content and historical data
Madison Cole Miner, Bellingham, WA (US); Eric Andrew Kolve, Seattle, WA (US); Jonathan Peter Manion, Bellingham, WA (US); and Robert Imre Gara, Jr., Seattle, WA (US)
Assigned to DexCare, Inc., Seattle, WA (US)
Filed by DexCare, Inc., Seattle, WA (US)
Filed on Mar. 25, 2024, as Appl. No. 18/615,623.
Int. Cl. G06Q 10/0639 (2023.01); G06F 40/40 (2020.01); G06Q 30/018 (2023.01); G16H 10/60 (2018.01); G16H 40/20 (2018.01); G16H 50/70 (2018.01)
CPC G16H 40/20 (2018.01) [G16H 10/60 (2018.01); G16H 50/70 (2018.01)] 28 Claims
OG exemplary drawing
 
1. A method for generating service offerings in a computing environment using one or more processors to execute instructions that are configured to cause actions, comprising:
determining content from a content panel based on one or more of a markup language, an encoding, or a format associated with the content panel;
determining one or more subjects associated with the content based on information included in the content and one or more evaluations of the content by a subject model;
determining a service category associated with the one or more subjects based on one or more services provided by a healthcare organization;
employing an offering model to generate an offering panel based on the service category and an availability of the one or more services, wherein the offering panel displays information associated with an available service, wherein the offering model includes a machine learning model;
evaluating the offering model based on monitoring one or more physical interactions between one or more of users and the offering panel;
employing one or more results of the evaluation to perform further actions, including:
designating the offering model for retraining based on the one or more performance metrics falling below a threshold value;
retraining the designated offering model based on one or more other metrics associated with one or more other offering models and a training model, wherein the training model includes one or more of a machine learning model or a large language model; and
employing the retrained offering model to generate one or more other offering panels for display to the one or more users; and
employing a deficiency model to compare the one or more physical interactions of the one or more users with the one or more offering panels to one or more previous physical interactions of the one or more users with one or more deficient offering panels; and
employing the comparison to determine one or more potential sources that improve relevance of the one or more offering panels for the one or more users and identify one or more causes of relevance deficiency for the one or more users.