Patent · US Active

Methods and apparatus for automated enterprise machine learning with data fused from multiple domains

US12400146B1 · kind B1 · utility

0Cited by
6References
20Claims
0Family size

Assignee

Inventors

Key dates

Filing dateApr 11, 2022
Grant dateAug 26, 2025
Priority date
Expiry dateDec 18, 2043

Classification

  • Technology area (CPC G)Physics
  • CPC primaryG06N20/00
  • WIPO fieldComputer technology
  • WIPO sectorElectrical engineering

Abstract

A method can include identifying a first set of data columns in a first data domain and a second set of data columns in a second data domain that are similar. The method can further include selecting a first subset of data columns in the first data domain and a second subset of data columns in the second data domain that have matching parameters. The method can further include arranging the first subset of data columns and the second set of data columns into a multi-domain dataset. Each column in the multi-domain dataset can be classified into structured data or unstructured data. The structured data can be used for training a machine learning model. The method can further include replacing under-performing features in the second data domain before replacing under-performing features in the first data domain. The method can further include re-training the machine learning model after replacing under-performing features.

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