Automated natural language communication analysis

Patent metadata

Field	Value
Publication number	US-11456979-B2
Application number	US-201916519841-A
Country	US
Kind code	B2
Filing date	Jul 23, 2019
Priority date	Jul 23, 2019
Publication date	Sep 27, 2022
Grant date	Sep 27, 2022

Abstract

A device may receive information identifying a communication framework for a mass communication task. The device may determine a success score for the communication framework using a mass communication model, wherein the success score represents a likelihood of a successful response in connection with using the communication framework for the mass communication task. The device may generate a recommendation for the communication framework based on the success score and using the mass communication model. The device may alter the communication framework to implement the recommendation and generate a modified communication framework. The device may perform the mass communication task using the modified communication framework.

First claim

What is claimed is:

1. A method, comprising:
receiving, by a device, information identifying a communication framework for a mass communication task;
determining, by the device, a success score for the communication framework using a mass communication model,
wherein the success score represents a likelihood of a successful response in connection with using the communication framework for the mass communication task;

generating, by the device, a recommendation for the communication framework based on the success score and using the mass communication model;
altering, by the device, the communication framework to implement the recommendation and generate a modified communication framework,
wherein altering the communication framework comprises:
optimizing an assignment of message recipients to time slots based on determining a ratio of responses received to messages sent in other mass communication tasks that includes a particular message recipient,
selecting, using the mass communication model and based on end-user data, a first timing for messaging associated with a first message recipient,
wherein selecting the first timing comprises:
 determining a first set of time slot scores for the first message recipient,
 wherein a time slot score, of the first set of time slot scores, represents a likelihood of response in a time slot of a set of time slots for messaging,
 determining a second set of time slot scores for the second message recipient, and
 optimizing an assignment of message recipients to time slots based on at least one of the first set of time slot scores, the second set of time slot scores, or a time slot availability criterion, and

selecting, using the mass communication model and based on the end-user data, a second timing that is different from the first timing for messaging associated with a second message recipient; and

performing, by the device, the mass communication task using the modified communication framework.

2. The method of claim 1 , further comprising:
receiving a mass communication data set identifying results of a set of mass communication tasks performed using a set of communication frameworks;
generating, using a machine learning technique and based on the mass communication data set, the mass communication model; and
storing, before receiving the information identifying the communication framework for the mass communication task, the mass communication model for subsequent use in evaluating the communication framework for the mass communication task.

3. The method of claim 1 , wherein the recommendation relates to at least one of:
a length of an initial message of the communication framework,
a relevance of the initial message of the communication framework,
a complexity of the communication framework,
a clarity of the communication framework,
a scheduling of messaging in accordance with the communication framework, or
a set of recipients of messages of the communication framework.

4. The method of claim 1 , wherein the success score corresponds to a likelihood of achieving a threshold response rate to one or more messages of the communication framework.

5. The method of claim 1 , wherein the mass communication model is trained using at least one of:
a random forest classifier technique,
a multilayer perceptron technique,
a stochastic gradient descent technique, or
a neural network technique.

6. The method of claim 1 , wherein the end-user data includes data identifying at least one of:
a message recipient location,
a message recipient job role,
a message recipient job level, or
a message recipient response history.

7. A device, comprising:
one or more memories; and
one or more processors communicatively coupled to the one or more memories, configured to:
receive a mass communication data set identifying results of a set of mass communication tasks performed using a set of communication frameworks;
generate, using a machine learning technique and based on the mass communication data set, a mass communication model;
store the mass communication model for subsequent use in evaluating a communication framework for a mass communication task;
receive, after storing the mass communication model, information identifying the communication framework for the mass communication task;
determine a success score for the communication framework using the mass communication model,
wherein the success score represents a likelihood of a successful response in connection with using the communication framework for the mass communication task;

generate a recommendation for the communication framework based on the success score and using the mass communication model;
alter the communication framework to implement the recommendation and generate a modified communication framework,
wherein, the one or more processors, when altering the communication framework, are to:
determine a ratio of responses received to messages sent in other mass communication tasks that includes a particular message recipient,
select, using the mass communication model and based on end-user data, a first timing for messaging associated with a first message recipient,
 wherein the one or more processors, to select the first timing, are configured to:
 determine a first set of time slot scores for the first message recipient,
 wherein a time slot score, of the first set of time slot scores, represents a likelihood of response in a time slot of a set of time slots for messaging,
 determine a second set of time slot scores for the second message recipient, and
 optimize an assignment of message recipients to time slots based on at least one of the first set of time slot scores, the second set of time slot scores, or a time slot availability criterion, and
 select, using the mass communication model and based on the end-user data, a second timing that is different from the first timing for messaging associated with a second message recipient; and

perform the mass communication task using the modified communication framework.

8. The device of claim 7 , wherein the one or more processors are configured to:
select, using the mass communication model and based on the end-user data, a first messaging channel for messaging associated with a third message recipient; and
select, using the mass communication model and based on the end-user data, a second messaging channel, that is different from the first messaging channel, for messaging associated with a fourth message recipient.

9. The device of claim 8 , wherein the end-user data includes data identifying at least one of:
a message recipient location,
a message recipient job role,
a message recipient job level, or
a message recipient response history.

10. The device of claim 8 , wherein the one or more processors, when selecting the first messaging channel, are configured to:
determine a first set of messaging channel scores for the first message recipient,
wherein a messaging channel score, of the first set of messaging channel scores, represents a likelihood of response in a messaging channel of a set of messaging channels for messaging;

determine a second set of messaging channel scores for the second message recipient; and
optimize an assignment of message recipients to messaging channels based on at least one of the first set of messaging channel scores, the second set of messaging channel scores, or a messaging channel availability criterion.

11. The device of claim 7 , wherein the one or more processors, when performing the mass communication task, are configured to:
transmit a set of messages to a set of message recipients;
monitor for a set of responses t

Assignees

• Accenture Global Solutions Ltd

Inventors

• Higgins Luke
• Kulkarni Aditi
• Vijayaraghavan Koushik M
• Donnelley Alastair
• Prasad Tanniru Rajendra

Classifications

• G06N20/00

  Machine learning · CPC title

• G06Q10/06393

  Score-carding, benchmarking or key performance indicator [KPI] analysis · CPC title

• H04L51/42

  Mailbox-related aspects, e.g. synchronisation of mailboxes · CPC title

• G06Q30/0203

  Market surveys; Market polls · CPC title

• G06Q10/10 Primary

  Office automation; Time management · CPC title