Patent · US Active

Knowledge graph question answering with neural machine translation

US12013884B2 · kind B2 · utility

0Cited by

1References

14Claims

0Family size

Assignee

International Business Machines Corporation · US

Inventors

Saswati Dana · Bengaluru, IN
Dinesh Garg · Beawar, IN
Dinesh Khandelwal · Indore, IN
G P Shrivatsa Bhargav · Bengaluru, IN
Sukannya Purkayastha · Dhubri, IN

Key dates

Filing date	Jun 30, 2022
Grant date	Jun 18, 2024
Priority date	—
Expiry date	Sep 15, 2042

Classification

Technology area (CPC G)Physics
CPC primaryG06N5/02
WIPO fieldComputer technology
WIPO sectorElectrical engineering

Abstract

A modular two-stage neural architecture is used in translating a natural language question into a logic form such as a SPARQL Protocol and RDF Query Language (SPARQL) query. In a first stage, a neural machine translation (NMT)-based sequence-to-sequence (Seq2Seq) model translates a question into a sketch of the desired SPARQL query called a SPARQL silhouette. In a second stage a neural graph search module predicts the correct relations in the underlying knowledge graph.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.