Back to results
US 12,754,783 B2Granted
Computer-based systems configured for a cloud-first multifunction API cluster providing microservices capable of both batch and real-time processing and method and use thereof
AssigneeCapital One Services, LLC
InventorsAli S. Al-Shehab, Jr-Wei Jeng, Niti N. Sheth, Tanveer Afzal Faruquie, David Edward Lutz, Nathan L. Sheridan
Life of this patent
Estimating the term…
Abstract
The disclosure is related to computer-based information processing systems configured to resolve entity records in a database. Specifically, the systems and methods utilize natural language models in a cloud-first multifunction API cluster to provide microservices capable of both batch and real-time processing of a large number of entity records.
Claims14
A computer-implemented method comprising:
- receiving, by at least one processor, a plurality of entity records, the plurality of entity records stored in an elastic search environment, the plurality of entity records associated with at least one candidate entity record;
- utilizing, by the at least one processor, in real time, a microservice module capable of leveraging a containerization technology in a cloud-first computer-implemented real-time scalable processing cluster to determine a match of each candidate entity pair, each candidate entity pair comprising:
- at least 20,000 entity records of the plurality of entity records and
- the at least one candidate entity record;
- wherein the microservice module monitoring the cloud-first computer-implemented real-time scalable processing cluster is configured to:
- utilize a cleansing engine to cleanse the at least one candidate entity record;
- determine a status of a blocking engine, the status representing a processing load associated with the blocking engine;
- utilize, responsive to the status of the blocking engine, the blocking engine to determine candidate entity pairs representing potential matches to the at least one candidate entity record from the plurality of entity records;
- utilize at least one feature engine to generate candidate entity pair features for each of the candidate entity pairs from the blocking engine based at least in part on the at least one candidate entity records and the potential matches;
- utilize at least one machine learning engine to determine a similarity score of each candidate entity pair from the blocking engine based at least in part on the candidate entity pair features of each candidate entity pair produced by the blocking engine;
- display the similarity score representing the at least one candidate entity record;
- determine at least one identified matching pair from the candidate entity pairs based at least in part on the similarity score of each candidate entity pair;
- and
- merge the at least one candidate entity record and the potential matches of the at least one identified match based on the similarity score and a user selection.
11 more elements
At least one non-transitory computer-readable storage medium having encoded thereon software instructions that, when executed by at least one processor, cause the at least one processor to perform steps to:
- receive, by at least one processor, a plurality of entity records, the plurality of entity records stored in an elastic search environment, the plurality of entity records associated with at least one candidate entity record;
- utilize, by the at least one processor, in real time, a microservice module capable of leveraging a containerization technology in a cloud-first computer-implemented real-time scalable processing cluster to determine a match of each candidate entity pair, each candidate entity pair comprising:
- at least 20,000 entity records of the plurality of entity records and
- the at least one candidate entity record;
- wherein the microservice module monitoring the cloud-first computer-implemented real-time scalable processing cluster is configured to:
- utilize a cleansing engine to cleanse the at least one candidate entity record;
- determine a status of a blocking engine, the status representing a processing load associated with the blocking engine;
- utilize, responsive to the status of the blocking engine, the blocking engine to determine candidate entity pairs representing potential matches to the at least one candidate entity record from the plurality of entity records;
- utilize at least one feature engine to generate candidate entity pair features for each of the candidate entity pairs from the blocking engine based at least in part on the at least one candidate entity records and the potential matches;
- utilize at least one machine learning engine to determine a similarity score of each candidate entity pair from the blocking engine based at least in part on the candidate entity pair features of each candidate entity pair produced by the blocking engine;
- display the similarity score representing the at least one candidate entity record;
- determine at least one identified matching pair from the candidate entity pairs based at least in part on the similarity score of each candidate entity pair;
- and
- merge the at least one candidate entity record and the potential matches of the at least one identified match based on the similarity score and a user selection.
11 more elements
Dependencies are inferred from the claim text.