Invention Grant
- Patent Title: Using relation suggestions to build a relational database
-
Application No.: US16106610Application Date: 2018-08-21
-
Publication No.: US11080300B2Publication Date: 2021-08-03
- Inventor: Alfio Massimiliano Gliozzo , Gaetano Rossiello
- Applicant: International Business Machines Corporation
- Applicant Address: US NY Armonk
- Assignee: International Business Machines Corporation
- Current Assignee: International Business Machines Corporation
- Current Assignee Address: US NY Armonk
- Agency: Cantor Colburn LLP
- Agent Anthony Curro
- Main IPC: G06F16/28
- IPC: G06F16/28 ; G06N5/00 ; G06F16/2455 ; G06F16/2457

Abstract:
Aspects of the invention include a system, a computer program product and a computer-implemented method for building a relational database between a first term and a second term. An embedding management system is used to determine a first term set based on the first term and a second term set based on the second term. An entity linking engine and a knowledge base engine are used to determine a relation match for the first term set when a relation of the first term set relates to a term from a second term set and for determining a relation match for the second term set when a relation of the second term set relates to a term from a first term set. A relation ranking engine selects a relation for the relational database having a selected number of matches.
Public/Granted literature
- US20200065416A1 USING RELATION SUGGESTIONS TO BUILD A RELATIONAL DATABASE Public/Granted day:2020-02-27
Information query