Invention Grant
- Patent Title: Systems and methods for generalized structured data discovery utilizing contextual metadata disambiguation via machine learning techniques
-
Application No.: US17010023Application Date: 2020-09-02
-
Publication No.: US11574129B2Publication Date: 2023-02-07
- Inventor: Santosh Chikoti , Jeffrey Kessler
- Applicant: JPMORGAN CHASE BANK, N.A.
- Applicant Address: US NY New York
- Assignee: JPMORGAN CHASE BANK, N.A.
- Current Assignee: JPMORGAN CHASE BANK, N.A.
- Current Assignee Address: US NY New York
- Agency: Greenberg Traurig LLP
- Main IPC: G06F40/30
- IPC: G06F40/30 ; G06F40/205 ; G06F16/2457 ; G06F40/253 ; G06N20/00

Abstract:
A method for generalized structured data discovery may include (1) receiving physical application metadata from data sources for an attribute, a database object, or a database; (2) receiving reference data comprising a plurality of tokens and their associated abbreviations/acronyms; (3) parsing the physical application metadata into a application tokens comprising known and unknown application tokens; (4) identifying unknown application tokens by comparing the parsed application tokens to a corpus; (5) performing probabilistic parsing on the unknown application tokens using the reference data; (6) performing bi-directional encoding to expand the polysemous tokens to relevant expressions using the reference data; (7) applying language tokens to the relevant expressions in the expanded polysemous tokens to disambiguate the relevant expressions; and (8) outputting a mapping of the physical application metadata to enhanced physical application metadata, wherein the enhanced physical application metadata comprises an expression for the physical application metadata in a supported language.
Public/Granted literature
Information query