Invention Grant
- Patent Title: Global explanations of machine learning model predictions for input containing text attributes
-
Application No.: US17535945Application Date: 2021-11-26
-
Publication No.: US11977836B1Publication Date: 2024-05-07
- Inventor: Cedric Philippe Archambeau , Sanjiv Ranjan Das , Michele Donini , Michaela Hardt , Tyler Stephen Hill , Krishnaram Kenthapadi , Pedro L Larroy , Xinyu Liu , Keerthan Harish Vasist , Pinar Altin Yilmaz , Muhammad Bilal Zafar
- Applicant: Amazon Technologies, Inc.
- Applicant Address: US WA Seattle
- Assignee: Amazon Technologies, Inc.
- Current Assignee: Amazon Technologies, Inc.
- Current Assignee Address: US WA Seattle
- Agency: Kowert, Hood, Munyon, Rankin & Goetzel, P.C.
- Agent Robert C. Kowert
- Main IPC: G06F7/02
- IPC: G06F7/02 ; G06F16/00 ; G06F16/22 ; G06F40/20 ; G06N5/01
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.
Information query
