公开(公告)号：US12118324B1

公开(公告)日：2024-10-15

申请号：US17710727

申请日：2022-03-31

Applicant: Amazon Technologies, Inc.

Inventor： Li Zhang ,  Sanjiv Ranjan Das ,  Yue Zhao ,  Zhijiang He ,  Shenghua Yue ,  Zheng Zhang ,  Xin Huang ,  Sheng Zha ,  Shuai Zheng

IPC: G06F16/35 ,  G06F16/34 ,  G06F40/284 ,  G06F40/40

CPC classification number: G06F40/40 ,  G06F16/345 ,  G06F16/355 ,  G06F40/284

Abstract: Techniques for machine learning (ML) and natural language processing (NLP) are described. One technique enables the creation of a clean training dataset through just a few API calls. Another technique provides an automated process for generating a domain-specific lexicon, which is then used to generate ML training datasets, in a manner that requires little to no human labor. Another technique gathers ML training data from domain-specific public sources, which are more likely than typical public sources to contain focused terminology and to be free from errors, thus resulting in trained ML models that provide more accurate inferences.

公开(公告)号：US11030394B1

公开(公告)日：2021-06-08

申请号：US15586778

申请日：2017-05-04

Applicant: Amazon Technologies, Inc.

Inventor： Zornitsa Petrova Kozareva ,  Sheng Zha ,  Hyokun Yun

IPC: G06N3/04 ,  G10L15/08 ,  G06F40/151 ,  G06N3/08 ,  G06F40/30

Abstract: A keyphrase extraction service implements techniques for determining a set of keyphrases associated with set of words. A word is selected from the set of words and a neural model is used to determine a label for the word based on features of the word and labels corresponding to other words of the set of words. The set of keyphrases is determined from the labels associated with the set of words.

公开(公告)号：US20240202458A1

公开(公告)日：2024-06-20

申请号：US18067680

申请日：2022-12-16

Applicant: Amazon Technologies, Inc.

Inventor： Sheng Zha ,  Miguel Ballesteros Martinez ,  Yassine Benajiba ,  Cole Hawkins ,  Aditya Rawal ,  Dhananjay Ram ,  Min Rong Samson Tan ,  Abhinav Goyal ,  Brant Swidler

IPC: G06F40/40 ,  G06F40/279

CPC classification number: G06F40/40 ,  G06F40/279 ,  G06F40/205

Abstract: Prompt discovery is performed for identifying prompts to natural language processing machine learning models. A request to determine a prompt for a natural language processing task performed by a pre-trained natural language processing machine learning model may be received. A task classification for the natural language processing task may be determined and candidate prompts for the natural language processing prompt task collection selected. Respective prompt results for the candidate prompts are evaluated to generate a prompt recommendation for the natural language processing task.

公开(公告)号：US20240202466A1

公开(公告)日：2024-06-20

申请号：US18067674

申请日：2022-12-16

Applicant: Amazon Technologies, Inc.

Inventor： Sheng Zha ,  Miguel Ballesteros Martinez ,  Yassine Benajiba ,  Cole Hawkins ,  Aditya Rawal ,  Dhananjay Ram ,  Min Rong Samson Tan ,  Vittorio Castelli

IPC: G06F40/56 ,  G06N5/04

CPC classification number: G06F40/56 ,  G06N5/04

Abstract: Prompt development techniques are implemented for tuning natural language processing machine learning models using selected prompts from a prompt task collection. A prompt development system may support requests to further adapt a pre-trained natural language processing machine learning model to tune the pre-trained natural language processing machine learning model for use with a selected prompt. Evaluation of the tuned natural language processing machine learning model may be performed and provided as a result.