公开(公告)号：US10826852B2

公开(公告)日：2020-11-03

申请号：US16177271

申请日：2018-10-31

Applicant: salesforce.com, inc.

Inventor： Zhao Jin ,  Shrestha Basu Mallick ,  Yacov Salomon ,  Kexin Xie ,  Sheng Loong Su ,  Todd Swardenski ,  Trent Albright ,  Armita Peymandoust ,  Michael Jones ,  Brian Brechbuhl ,  David Yourdon

IPC: H04L12/58 ,  H04L29/08

Abstract: A database server may receive or monitor user engagement metadata corresponding to a plurality of communication messages transmitted to the users. The database server analyzes the metadata to determine optimal transmission frequencies for digital communication messages based on engagement rates received in the user engagement metadata.

公开(公告)号：US20200137008A1

公开(公告)日：2020-04-30

申请号：US16177271

申请日：2018-10-31

Applicant: salesforce.com, Inc.

Inventor： Zhao Jin ,  Shrestha Basu Mallick ,  Yacov Salomon ,  Kexin Xie ,  Sheng Loong Su ,  Todd Swardenski ,  Trent Albright ,  Armita Peymandoust ,  Michael Jones ,  Brian Brechbuhl ,  David Yourdon

IPC: H04L12/58 ,  H04L29/08

Abstract: A database server may receive or monitor user engagement metadata corresponding to a plurality of communication messages transmitted to the users. The database server analyzes the metadata to determine optimal transmission frequencies for digital communication messages based on engagement rates received in the user engagement metadata.

公开(公告)号：US20180101521A1

公开(公告)日：2018-04-12

申请号：US15840482

申请日：2017-12-13

Applicant: salesforce.com, inc.

Inventor： Michael Jones ,  Scott Enman ,  Collin Chun-Kit Lee ,  David Campbell ,  Christopher John Nicholls

IPC: G06F17/27 ,  G06F17/21 ,  G06Q50/00 ,  G06F17/24

CPC classification number: G06F17/2785 ,  G06F17/218 ,  G06F17/24 ,  G06Q50/01

Abstract: Provided are techniques for avoiding sentiment model overfitting in a machine language model. A current list of keywords in a current sentiment model can be updated to create a proposed list of keywords in a proposed sentiment model. Machine-generated sentiment results, based on the proposed sentiment model, are presented to identify model overfitting, without revising the current set of keywords. The proposed set of keywords can be edited, and when overfitting is not present, the current list of keywords is replaced by the proposed list of keywords.

公开(公告)号：US09881000B1

公开(公告)日：2018-01-30

申请号：US15213158

申请日：2016-07-18

Applicant: SALESFORCE.COM, INC.

Inventor： Michael Jones ,  Scott Enman ,  Collin Chun-Kit Lee ,  David Campbell ,  Christopher John Nicholls

IPC: G06F17/27 ,  G06F17/28 ,  G06F17/21 ,  G06Q50/00 ,  G06F17/24

CPC classification number: G06F17/2785 ,  G06F17/218 ,  G06F17/24 ,  G06Q50/01

Abstract: Provided are techniques for avoiding sentiment model overfitting in a machine language model. A current list of keywords in a current sentiment model can be updated to create a proposed list of keywords in a proposed sentiment model. Machine-generated sentiment results, based on the proposed sentiment model, are presented to identify model overfitting, without revising the current set of keywords. The proposed set of keywords can be edited, and when overfitting is not present, the current list of keywords is replaced by the proposed list of keywords.