Patent · US Active

Global explanations of machine learning model predictions for input containing text attributes

US11977836B1 · kind B1 · utility

0Cited by

0References

20Claims

0Family size

Assignee

AMAZON TECHNOLOGIES, INC. · US

Inventors

Cedric Philippe Archambeau · Berlin, DE
Sanjiv Ranjan Das · San Jose, US
Michele Donini · Berlin, DE
Michaela Hardt · Berkeley, US
Tyler Stephen Hill · Los Altos, US
Krishnaram Kenthapadi · Sunnyvale, US
Pedro Larroy · Berlin, DE
Xinyu Liu · Nanhu, CN
Keerthan Harish Vasist · Sunnyvale, US
Pinar Altin Yilmaz · Palo Alto, US
Muhammad Bilal Zafar · Berlin, DE

Key dates

Filing date	Nov 26, 2021
Grant date	May 7, 2024
Priority date	—
Expiry date	Oct 22, 2042

Classification

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

Abstract

A determination is made that an explanatory data set for a common set of predictions generated by a machine learning model for records containing text tokens is to be provided. Respective groups of related tokens are identified from the text attributes of the records, and record-level prediction influence scores are generated for the token groups. An aggregate prediction influence score is generated for at least some of the token groups from the record-level scores, and an explanatory data set based on the aggregate scores is presented.

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