Granted Patent B1

US 11,768,837 · App. 17/646,228 · Granted Sep 26, 2023

Inventors: David Newman (Walnut Creek, CA); Omar B. Khan (Richmond, VA); Alexander Joseph Kalinowski (Philadelphia, PA)

Assignee: Wells Fargo Bank, N.A.

G06F16/24575G06F16/2423G06N5/02

Abstract

Claims (61)

1. A computer-implemented method comprising:

storing a vector space representation of a set of concepts in a semantic ontology, wherein locations of the set of concepts in the vector space are based on weights of a neural network trained using triples from the semantic ontology;

presenting a semantic search user interface, the semantic search user interface including:

a text input portion to receive a search query;

a suggested search portion; and

a search results portion;

receiving, from a user computing device, the search query;

converting, using at least one processor, the search query into a vector;

computing, using the at least one processor, a set of possible concepts related to the search query based on a distance between the vector and other concepts in the vector space representation;

presenting the set of possible concepts related to the search query in the suggested search portion of the semantic search user interface;

receiving, from the user device, a selection of a concept in the set of possible concepts;

in response to receiving the selection, executing a query to a graph database to retrieve concepts semantically linked to the concept; and

presenting the concepts semantically linked to the concept in the search results portion of the semantic search user interface.

2. The computer-implemented method of claim 1 , wherein converting, using at least one processor, the search query into the vector includes:

matching the search query to a concept in the semantic ontology; and

performing a lookup in an index for the concept to retrieve the vector.

3. The computer-implemented method of claim 2 , wherein matching the search query to a concept in the semantic ontology includes correcting, using the at least one processor, spelling of a word in the search query.

4. The computer-implemented method of claim 1 , wherein the set of possible concepts includes a direct match to a concept in the set of concepts in the semantic ontology.

5. The computer-implemented method of claim 1 , wherein computing the set of possible concepts related to the search query based on the distance between the vector and other concepts in the vector space representation includes calculating cosign similarity values between the vector and respective vectors of the other concepts.

6. The computer-implemented method of claim 1 , wherein the concepts semantically linked to the concept include properties of the concept and data assets storing data associated with the concept.

7. A system comprising:

at least one processor; and

a storage device comprising instructions, which when executed by the at least one processor, configure the at least one processor to perform operations comprising:

storing a vector space representation of a set of concepts in a semantic ontology, wherein locations of the set of concepts in the vector space are based on weights of a neural network trained using triples from the semantic ontology;

presenting a semantic search user interface, the semantic search user interface including:

a text input portion to receive a search query;

a suggested search portion; and

a search results portion;

receiving, from a user computing device, the search query;

converting the search query into a vector;

computing a set of possible concepts related to the search query based on a distance between the vector and other concepts in the vector space representation;

presenting the set of possible concepts related to the search query in the suggested search portion of the semantic search user interface;

receiving, from the user device, a selection of a concept in the set of possible concepts;

in response to receiving the selection, executing a query to a graph database to retrieve concepts semantically linked to the concept; and

presenting the concepts semantically linked to the concept in the search results portion of the semantic search user interface.

8. The system of claim 7 , wherein converting the search query into the vector includes:

matching the search query to a concept in the semantic ontology; and

performing a lookup in an index for the concept to retrieve the vector.

9. The system of claim 8 , wherein matching the search query to a concept in the semantic ontology includes correcting spelling of a word in the search query.

10. The system of claim 7 , wherein the set of possible concepts includes a direct match to a concept in the set of concepts in the semantic ontology.

11. The system of claim 7 , wherein computing the set of possible concepts related to the search query based on the distance between the vector and other concepts in the vector space representation includes calculating cosign similarity values between the vector and respective vectors of the other concepts.

12. The system of claim 7 wherein the concepts semantically linked to the concept include properties of the concept and data assets storing data associated with the concept.

13. A non-transitory computer-readable medium comprising instructions, which when executed by at least one processor, configure the at least one processor to perform operations comprising:

storing a vector space representation of a set of concepts in a semantic ontology, wherein locations of the set of concepts in the vector space are based on weights of a neural network trained using triples from the semantic ontology;

presenting a semantic search user interface, the semantic search user interface including:

a text input portion to receive a search query;

a suggested search portion; and

a search results portion;

receiving, from a user computing device, the search query;

converting the search query into a vector;

computing a set of possible concepts related to the search query based on a distance between the vector and other concepts in the vector space representation;

presenting the set of possible concepts related to the search query in the suggested search portion of the semantic search user interface;

receiving, from the user device, a selection of a concept in the set of possible concepts;

in response to receiving the selection, executing a query to a graph database to retrieve concepts semantically linked to the concept; and

presenting the concepts semantically linked to the concept in the search results portion of the semantic search user interface.

14. The non-transitory computer-readable medium of claim 13 , wherein converting the search query into the vector includes:

matching the search query to a concept in the semantic ontology; and

performing a lookup in an index for the concept to retrieve the vector.

15. The non-transitory computer-readable medium of claim 14 , wherein matching the search query to a concept in the semantic ontology includes correcting spelling of a word in the search query.

16. The non-transitory computer-readable medium of claim 13 , wherein the set of possible concepts includes a direct match to a concept in the set of concepts in the semantic ontology.

17. The non-transitory computer-readable medium of claim 13 , wherein computing the set of possible concepts related to the search query based on the distance between the vector and other concepts in the vector space representation includes calculating cosign similarity values between the vector and respective vectors of the other concepts.

Assignments (2)

Cited By (10)

US 12,204,509 US 12,212,575 US 12,242,491 US 12,306,844 US 12,314,304 US 12,314,347 US 12,326,867 US 12,335,341 US 12,339,861 US 12,705,268