Granted Patent B2

US 12,443,856 · App. 16/731,965 · Granted Oct 14, 2025

Inventors: Parag Goradia (San Ramon, CA); Ryan Andrew Oattes (Burlington, CA); Lakshmi Prasad Reddy Nanga (Pleasanton, CA); Luis Antonio Monterrubio Ruiz (Burlington, CA)

Assignee: Kobai, Inc.

G06N5/02G06F16/252G06N20/00

Abstract

Claims (34)

1. A method of providing decision intelligence, the method comprising:

capturing a business knowledge model from an expert user to have a semantic ontology, the business knowledge model being associated with a business question and including concepts and properties;

mapping data for the business knowledge model, wherein the mapping comprises:

receiving, via a user interface, identification of a first column in a first table, wherein the first column comprises property values;

based on the receiving the identification of the first column in the first table: generating a prompt to indicate at least one second column to uniquely identify concept instances to which the properties belong;

in response to the prompt, receiving, via the user interface, identification of the at least one second column in the first table;

identifying a function to be applied to values of the at least one second column, wherein each respective adapted value is derived based on applying the indicated identified function to the respective value of the at least one second column and uniquely identifies a concept instance with a concept property matching the respective property value of the property values of the first column;

identifying a plurality of concept instances in the business knowledge model, wherein each respective concept instance matches the respective value derived based on applying the identified function to the respective value of the at least one second column; and

updating each respective concept instance of the first plurality of concept instances based on a respective property value in the first column of the first table; and

generating classification tags for access control;

applying one or more classification tags to each property of the properties of the plurality of concept instances;

assigning one or more classification tags to each user of a plurality of users;

generating an application programming interface (API) to deploy the business knowledge model and the associated business question, concepts, and properties; and

implementing access control for the API in which an access control condition requires a user of the plurality of users to have assigned classification tags that match all of the one or more classification tags that are applied to the property of an answer provided using the API to the associated business question for the user to be granted access to the answer to the business question.

2. The method of claim 1 , wherein the plurality of concept instances is a first plurality of concept instances and wherein the mapping of data further comprises:

receiving, via the user interface, identification of a third column in a second table, wherein the third column comprises property values;

receiving, via the user interface, identification of a fourth column in the second table, wherein the fourth column comprises unique property values, and wherein the data in the fourth column of the second table is of a different data type than data in the at least one second column of the first table;

identifying a second plurality of concept instances in the business knowledge model, wherein each respective concept instance matches a respective unique property value in the fourth column of the second table; and

updating each respective concept instance of the second plurality of concept instances based on a respective property value in the third column of the second table.

3. The method of claim 2 , wherein the mapping of data further comprises adapting to differences in source data using at least one adaptation function.

4. The method of claim 2 , wherein the mapping of data further comprises mapping a concept to two or more source data tables.

5. The method of claim 4 , wherein a hierarchy is used to map a concept to the two or more source data tables.

6. The method of claim 2 , wherein the mapping of data further comprises mapping data with disjoint identifiers.

7. The method of claim 1 , further wherein the business question and the business knowledge model are bundled into a solution as an instantiation of the business knowledge model and the business question.

8. The method of claim 1 , wherein the data mapping links the business knowledge model to underlying data.

9. The method of claim 1 , wherein the business knowledge model is translated into the semantic ontology.

10. The method of claim 1 , wherein the API is deployed in response to a single click of the user interface.

11. The method of claim 1 , wherein the business knowledge model is generated independently of the data structure of at least one data source used to populate the business knowledge model with data.

12. The method of claim 1 , wherein the mapping comprises:

generating a time trigger based on the data;

selecting a subset of time series data based on the generated time trigger; and

mapping at least one value generated based on the selected subset of time series data to a concept instance of the business knowledge model.

13. The method of claim 12 , wherein selecting the subset of the time series data based on the generated time trigger comprises selecting a predetermined number of entries in the time series data before or after the generated time trigger.

14. The method of claim 13 , further comprising utilizing the time series calculation for asset relationship management.

Assignments (2)

Continuity (2)

References Cited (60)