Granted Patent B2

US 12,731,189 · App. 19/205,688 · Granted Sep 8, 2026

Inventors: Devon Drew (Houston, TX); Kevin D. Howard (Mesa, AZ)

Assignee: Diligence Fund Distributors Inc.

G06Q40/06

Abstract

Claims (49)

1 . A method for optimizing financial decision making, the method comprising:

receiving, financial context data associated with a financial decision;

receiving unstructured data from a plurality of data sources;

processing the unstructured data to generate a set of event features and sentiment features using a trained model, the trained model performing natural language processing to derive semantic embeddings for the sentiment features;

determining, based at least in part on the financial context data, the event features, and the sentiment features, predictions for the financial decision;

generating a list of workflow options based at least in part on the predictions for the financial decision;

selecting a workflow option from the list of workflow options; and

returning information about the selected workflow option for processing related to the financial decision.

2 . The method of claim 1 , wherein the financial context data comprises at least one of investments data, funds data, investment types data, or qualifying investments.

3 . The method of claim 1 , further comprising:

rating each workflow option in the list of workflow options; and

selecting the workflow option from the list of workflow options based on the respective rating.

4 . The method of claim 3 , further comprising:

selecting a portion of the list of workflow options based on respective ratings; and

presenting the portion of the list of workflow options to a client.

5 . The method of claim 4 , further comprising:

receiving client selection of the workflow option; and

performing an action based on the client-selected workflow option.

6 . The method of claim 1 , wherein returning the information comprises performing actions according to the selected workflow option.

7 . The method of claim 1 , wherein the financial decision comprises identifying an investment opportunity.

8 . The method of claim 1 , wherein the unstructured data comprises at least social media content and newsfeed data.

9 . The method of claim 1 , further comprising automatically executing a workflow corresponding to the selected workflow option.

10 . The method of claim 9 , wherein the workflow defines a sequence of operations to be performed that relate to the financial decision.

11 . The method of claim 1 , wherein processing the unstructured data to generate the set of event features and sentiment features using the trained model comprises:

extracting text strings from the unstructured data;

converting the text strings into a list of tokens; and

identifying phrases and words from the list of tokens that are associated with a particular subject to generate the set of event features and sentiment features.

12 . The method of claim 11 , wherein the trained model comprises a bidirectional encoder representations from transformer (BERT) model.

13 . A non-transitory computer-readable medium storing instructions that, when executed by one or more processors of a computing system, cause the computing system to perform operations comprising:

receiving, financial context data associated with a financial decision;

receiving unstructured data from a plurality of data sources;

processing the unstructured data to generate a set of event features and sentiment features using a trained model, the trained model performing natural language processing to derive semantic embeddings for the sentiment features;

determining, based at least in part on the financial context data, the event features, and the sentiment features, predictions for the financial decision;

generating a list of workflow options based at least in part on the predictions for the financial decision;

selecting a workflow option from the list of workflow options; and

returning information about the selected workflow option for processing related to the financial decision.

14 . The non-transitory computer-readable medium of claim 13 , wherein the financial context data comprises at least one of investments data, funds data, investment types data, or qualifying investments.

15 . The non-transitory computer-readable medium of claim 13 , further comprising:

rating each workflow option in the list of workflow options; and

selecting the workflow option from the list of workflow options based on the respective rating.

16 . The non-transitory computer-readable medium of claim 15 , further comprising:

selecting a portion of the list of workflow options based on respective ratings; and

presenting the portion of the list of workflow options to a client.

17 . The non-transitory computer-readable medium of claim 16 , further comprising:

receiving client selection of the workflow option; and

performing an action based on the client-selected workflow option.

18 . The non-transitory computer-readable medium of claim 13 , wherein returning the information comprises performing actions according to the selected workflow option.

19 . The non-transitory computer-readable medium of claim 13 , further comprising automatically executing a workflow corresponding to the selected workflow option.

20 . The non-transitory computer-readable medium of claim 19 , wherein the workflow defines a sequence of operations to be performed that relate to the financial decision.

Assignments (1)

Continuity (2)

References Cited (82)