Hyper-segmented personalization using machine learning-based models in an information processing system

Patent metadata

Field	Value
Publication number	US-12217297-B2
Application number	US-202217746034-A
Country	US
Kind code	B2
Filing date	May 17, 2022
Priority date	May 17, 2022
Publication date	Feb 4, 2025
Grant date	Feb 4, 2025

Abstract

Techniques are disclosed for hyper-segmented personalization using machine learning-based models in an information processing system. For example, a method obtains one or more product experience recommendation data sets respectively from one or more product entities, and one or more purchase experience recommendation data sets respectively from one or more commerce entities. The method applies a federated ensemble-based machine learning algorithm to at least one of the one or more purchase experience recommendation data sets and at least one of the one or more product experience recommendation data sets to generate a personalized model, and causes adaptation of a purchasing interface of at least one of the one or more commerce entities with respect to a given user based on the personalized model.

First claim

What is claimed is:

1. An apparatus comprising:
a processing platform comprising at least one processor coupled to at least one memory, the processing platform, when executing program code, being configured to:
identify, using one or more machine learning models, one or more similarities between product experience data of a given user and product experience data of one or more users to generate one or more product experience recommendations for the given user,
identify, using the one or more machine learning models, one or more similarities between purchase experience data of the given user and purchase experience data of the one or more users to generate one or more purchase experience recommendations for the given user;
generate a first machine learning model comprising at least one or more product experience recommendation data sets respectively from one or more product entities, wherein each of the one or more product experience recommendation data sets corresponds to one or more products produced by a respective one of the one or more product entities and is based on the one or more product experience recommendations for the given user with respect to at least one of the one or more products;
generate a second machine learning model comprising at least one or more purchase experience recommendation data sets respectively from one or more commerce entities, wherein each of the one or more purchase experience recommendation data sets corresponds to the one or more products sold by a respective one of the one or more commerce entities and is based on the one or more purchase experience recommendations for the given user with respect to at least one of the one or more products;
apply a federated ensemble-based machine learning algorithm to a ground truth label, at least one of the one or more purchase experience recommendation data sets and at least one of the one or more product experience recommendation data sets to generate and train a personalized model, wherein the application of the federated ensemble-based machine learning algorithm comprises:
aggregating the first machine learning model and the second machine learning model into the personalized model by combining each of the one or more product experience recommendation data sets with each of the one or more purchase experience recommendation data sets; and
determining one or more adaptations to be implemented on a purchasing interface of at least one of the one or more commerce entities with respect to a given one of the one or more users based on the personalized model to recommend one or more products to the given one of the one or more users; and

cause the one or more adaptations to be maintained wherein the given one of the one or more users is enabled to switch from the purchasing interface of at least one of the one or more commercial entities to another purchasing interface of at least another of the one or more commercial entities.

2. The apparatus of claim 1 , wherein each of the one or more product experience recommendation data sets comprises a machine learning model of the one or more machine learning models.

3. The apparatus of claim 1 , wherein each of the one or more product experience recommendation data sets comprises one or more recommendations based on one or more of ratings data and behavior data collected from at least a portion of the one or more users.

4. The apparatus of claim 1 , wherein each of the one or more purchase experience recommendation data sets comprises a machine learning model of the one or more machine learning models.

5. The apparatus of claim 1 , wherein each of the one or more purchase experience recommendation data sets comprises one or more recommendations based on one or more of product data and support data collected from at least a portion of the one or more users.

6. The apparatus of claim 1 , wherein applying a federated ensemble-based machine learning algorithm further comprises:
aggregating the at least one of the one or more purchase experience recommendation data sets and the at least one of the one or more product experience recommendation data sets to form the personalized model to distill knowledge from both the at least one of the one or more purchase experience recommendation data sets and the at least one of the one or more product experience recommendation data sets into the personalized model.

7. The apparatus of claim 6 , wherein forming the personalized model further comprises:
predicting recommendation data for one or more unavailable product experience recommendation data sets; and
aggregating the predicted recommendation data with one or more available product experience recommendation data sets.

8. The apparatus of claim 6 , wherein forming the personalized model further comprises:
computing probabilities associated with the at least one of the one or more purchase experience recommendation data sets and the at least one of the one or more product experience recommendation data sets to generate the personalized model.

9. The apparatus of claim 8 , wherein computing probabilities associated with the at least one of the one or more purchase experience recommendation data sets and the at least one of the one or more product experience recommendation data sets to generate the personalized model further comprises:
applying a Bayesian statistical model to the at least one of the one or more purchase experience recommendation data sets and the at least one of the one or more product experience recommendation data sets.

10. The apparatus of claim 1 , wherein causing adaptation of a purchasing interface of at least one of the one or more commerce entities with respect to a given one of the one or more users based on the personalized model further comprises:
causing presentation of one or more recommendations on the purchasing interface that are hyper-personalized for the given user.

11. The apparatus of claim 1 , wherein causing adaptation of a purchasing interface of at least one of the one or more commerce entities with respect to a given one of the one or more users based on the personalized model further comprises:
causing presentation of the one or more products on the purchasing interface consistent with one or more preferences of the corresponding one of the one or more product entities associated with the at least one of the one or more product experience recommendation data sets.

12. A method comprising:
identifying, using one or more machine learning models, one or more similarities between product experience data of a given user and product experience data of one or more users to generate one or more product experience recommendations for the given user;
identifying, using the one or more machine learning models, one or more similarities between purchase experience data of the given user and purchase experience data of the one or more users to generate one or more purchase experience recommendations for the given user;
generating a first machine learning model comprising at least one or more product experience recommendation data sets respectively from one or more product entities, wherein each of the one or more product experience recommendation data sets corresponds to one or more products produced by a respective one of the one or more product entities and is based on the one or more product experience recommendations for the given user with respect to at least one of the one or more products;
generating a second machine learning model comprising at least one or more purchase experience recommendation data sets respectively from one or more commerce entities, wherein each of the one or more purchase experience recommendation data s

Assignees

• Dell Products Lp

Inventors

• Kumar Dhilip S
• Sahoo Sujit Kumar
• Maikhuri Ajay
• Marappa Devaraj

Classifications

• G06N7/01

  Probabilistic graphical models, e.g. probabilistic networks · CPC title

• G06N20/20

  Ensemble learning · CPC title

• G06N3/098

  Distributed learning, e.g. federated learning · CPC title

• G06Q30/0631 Primary

  Recommending goods or services · CPC title