US 11,107,115 B2Granted

Machine-learning based multi-step engagement strategy modification

AssigneeAdobe Inc.

InventorsPankhri Singhai, Sundeep Parsa, Piyush Gupta, Nupur Kumari, Nikaash Puri, Mayank Singh, Eshita Shah, Balaji Krishnamurthy, Akash Rupela

Life of this patent

Estimating the term…

patentgrep

Loading patent…

Abstract

Machine-learning based multi-step engagement strategy modification is described. Rather than rely heavily on human involvement to manage content delivery over the course of a campaign, the described learning-based engagement system modifies a multi-step engagement strategy, originally created by an engagement-system user, by leveraging machine-learning models. In particular, these leveraged machine-learning models are trained using data describing user interactions with delivered content as those interactions occur over the course of the campaign. Initially, the learning-based engagement system obtains a multi-step engagement strategy created by an engagement-system user. As the multi-step engagement strategy is deployed, the learning-based engagement system randomly adjusts aspects of the sequence of deliveries for some users. Based on data describing the interactions of recipients with deliveries served according to both the user-created and random multi-step engagement strategies, the machine-learning models generate a modified multi-step engagement strategy.

Claims20

Dependent claims

A computer-implemented method comprising:

receiving, by at least one computing device, a user-created multi-step engagement strategy for controlling delivery of content of a campaign, the user-created multi-step engagement strategy including an entry condition, an exit condition, and at least two content deliveries to each of a plurality of client device users targeted by the campaign based on the entry condition, the entry condition, the exit condition, and the at least two content deliveries of the user-created multi-step engagement strategy having been specified by a user via a user interface;
generating, by the at least one computing device, at least one random multi-step engagement strategy for controlling the delivery of the content by randomly adjusting at least one aspect of how the at least two content deliveries deliver the content to targeted client device users;
delivering, by the at least one computing device, the content of the campaign to the targeted client device users according to both the user-created and at least one random multi-step engagement strategies;
generating, by the at least one computing device, a prediction of a multi-step engagement strategy that optimizes performance of the exit condition across the targeted client device users using a machine-learning model, the machine-learning model trained to optimize the performance of the exit condition with data describing interactions of the targeted client device users with the content delivered according to both the user-created and at least one random multi-step engagement strategies;
and
modifying, by the at least one computing device, the delivery of the content based on the prediction of the multi-step engagement strategy.

3 more elements

A system comprising:

a content delivery module implemented at least partially in hardware of at least one computing device to deliver content of a campaign according to multi-step engagement strategies;
a machine-learning model implemented at least partially in the hardware of the at least one computing device to generate a prediction of a multi-step engagement strategy that optimizes performance of a campaign exit condition across client device users targeted by the campaign;
and
an engagement-strategy manager module implemented at least partially in the hardware of the at least one computing device to:
generate at least one random multi-step engagement strategy for controlling delivery of the content of the campaign by randomly adjusting at least one aspect of how at least two content deliveries of a user-created multi-step engagement strategy deliver the content to targeted client device users, the at least two content deliveries of the user-created multi-step engagement strategy having been specified by a user via a user interface;
train the machine-learning model to generate a prediction of a multi-step engagement strategy based on data describing interactions of the targeted client device users with the content delivered according to both the user-created and at least one random multi-step engagement strategies;
and
determine an amount of the content to deliver according to the prediction of the multi-step engagement strategy.

5 more elements

One or more computer-readable storage media having stored thereon instructions that are executable by one or more processors to perform operations comprising:

receiving a user-created multi-step engagement strategy for controlling delivery of content of a campaign, the user-created multi-step engagement strategy including an entry condition, an exit condition, and at least two content deliveries to each of a plurality of client device users targeted by the campaign based on the entry condition, the entry condition, the exit condition, and the at least two content deliveries of the user-created multi-step engagement strategy having been specified by a user via a user interface;
generating at least one random multi-step engagement strategy for controlling the delivery of the content by randomly adjusting at least one aspect of how the at least two content deliveries deliver the content to targeted client device users;
delivering the content of the campaign to the targeted client device users according to both the user-created and at least one random multi-step engagement strategies;
generating a prediction of a multi-step engagement strategy that optimizes performance of the exit condition across the targeted client device users using a machine-learning model, the machine-learning model trained to optimize the performance of the exit condition with data describing interactions of the targeted client device users with the content delivered according to both the user-created and at least one random multi-step engagement strategies;
and
modifying the delivery of the content based on the prediction of the multi-step engagement strategy.

3 more elements

Dependencies are inferred from the claim text.