Multi-stage content analysis system that profiles users and selects promotions

• US 10,528,984 B2
• Filed: 02/11/2019
• Issued: 01/07/2020
• Est. Priority Date: 10/24/2017
• Status: Active Grant

• Current Assignee: Kaptivating Technology LLC
• Original Assignee: Kaptivating Technology LLC
• Inventors: Palmer, Michael, Tulk, Steven, Galido, Thomas
• Primary Examiner(s): Summers, Kiersten V
• Application Number:
  US16/272,790
• Publication Number:
  US 20190188758A1
• Time in Patent Office: 330 Days
• Field of Search: 705 1452, 705 1453, 705 1466
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks
• CPC Class Codes: G06Q 30/0254 based on statistics
• CPC Class Codes: G06Q 30/0255 based on user history
• CPC Class Codes: G06Q 30/0256 User search
• CPC Class Codes: G06Q 30/0269 based on user profile or attribute
• CPC Class Codes: G06Q 30/0277 Online advertisement

Abstract

A system that analyzes a user'"'"'s communications to select a promotion that is presented to the user. The analysis may occur in two stages: a first stage analyzes a single communication from a user to determine whether the user is a potential target for a promotion; for potential targets, a second stage analyzes a history of communications from the user to generate a user profile. The system may then select a promotion based on the profile. The profile may include a set of profile tags that are considerably more detailed and granular than traditional demographic data; tags may for example indicate user affiliations with groups or ideas (such as religions or political parties), or user life cycle stages. Using these rich, detailed user profile tags, the system may achieve promotion response rates far above those from traditional advertising, which relies on cookies or simple demographic categories.

Claims

1. A multi-stage content analysis system that executes on a computer that profiles users and selects promotions, comprising:

a first analysis stage configured to

receive a communication created by a user; and

,

analyze said communication to determine whether said user is a potential target for one or more promotions;

a second analysis stage coupled to said first analysis stage and configured to

when said first analysis stage determines that said user is a potential target for a promotion

receive a communications history associated with said user, wherein said communications history comprises a plurality of communications created by said user; and

,

analyze said communications history to assign one or more user profile tags to said user, wherein said analyze said communications history to assign said one or more user profile tags to said user comprises;

access a database comprising key words and phrases associated with each tag of a set of one or more tags;

calculate a frequency of each of said key words and phrases in said communications history;

calculate a tag relevance score for each tag of said set of one or more tags based on said frequency of each of said key words and phrases, wherein said calculate said tag relevance score for each tag of said set of one or more tags comprises;

calculate a probability that said communications history is associated with each tag of said set of one or more tags using a naï

ve Bayes classifier, wherein said frequency of each of said key words and phrases in said communications history comprises a feature vector for said naï

ve Bayes classifier; and

,

a promotion selector coupled to said second analysis stage and configured to

receive said one or more user profile tags from said second analysis stage;

analyze said one or more user profile tags and said communication to select a specific promotion from said one or more promotions; and

,

transmit said specific promotion to said user.

2. The system of claim 1, wherein said one or more user profile tags comprise one or more affiliations of said user with an organization, group, cause, or belief.

3. The system of claim 2, wherein said organization, group, cause, or belief comprises a religion.

4. The system of claim 2, wherein said organization, group, cause, or belief comprises a political party, a political viewpoint, or a political candidate.

5. The system of claim 1, wherein said one or more user profile tags comprise one or more life cycle stages associated with said user.

6. The system of claim 5, wherein said one or more life cycle stages comprise one or more of teenager, prospective student, student, new graduate, early adult, expectant spouse, new spouse, expectant new parent, new parent, expectant empty nester, empty nester, senior, grandparent, expectant retiree, and retiree.

7. The system of claim 1, further comprising:

a third analysis stage coupled to said second analysis stage and configured to

access an external database of user profile information;

modify or extend said one or more user profile tags based on said external database.

8. The system of claim 1, further comprising:

a machine learning engine coupled to one or more of said first analysis stage, said second analysis stage, and said promotion selector, and configured to

receive data describing one or more of

whether one or more users responded to one or more promotions that were transmitted to said one or more users;

how said one or more users responded to said one or more promotions; and

,

purchases, subscriptions, or enrollments made by said one or more users; and

,

execute a machine learning algorithm on said data to update one or more of said first analysis stage, said second analysis stage, and said promotion selector.

9. The system of claim 8, wherein said analyze said communications history to assign said one or more user profile tags to said user comprises:

said machine learning algorithm is configured to modify said key words and phrases associated with one or more of said set of one or more tags.

10. A multi-stage content analysis system that profiles users and selects promotions, comprising:

a first analysis stage configured to

receive a communication created by a user; and

,

analyze said communication to determine whether said user is a potential target for one or more promotions;

a second analysis stage coupled to said first analysis stage and configured to

when said first analysis stage determines that said user is a potential target for a promotion

receive a communications history associated with said user, wherein said communications history comprises a plurality of communications created by said user; and

,

analyze said communications history to assign one or more user profile tags to said user;

wherein said analyze said communications history to assign one or more user profile tags to said user comprises;

access a database comprising key words and phrases associated with each tag of a set of one or more tags;

calculate a frequency of each of said key words and phrases in said communications history; and

,

calculate a probability that said communications history is associated with each tag of said set of one or more tags using a naï

ve Bayes classifier, wherein said frequency of each of said key words and phrases in said communications history comprises a feature vector for said naï

ve Bayes classifier;

a promotion selector coupled to said second analysis stage and configured to

receive said one or more user profile tags from said second analysis stage;

analyze said one or more user profile tags and said communication to select a specific promotion from said one or more promotions; and

,

transmit said specific promotion to said user; and

,

a machine learning engine coupled to one or more of said first analysis stage, said second analysis stage, and said promotion selector, and configured to

receive data describing one or more of

whether one or more users responded to said one or more promotions that were transmitted to said one or more users;

how said one or more users responded to said one or more promotions; and

,

purchases, subscriptions, or enrollments made by said one or more users; and

,

execute a machine learning algorithm on said data to update one or more of said first analysis stage, said second analysis stage, and said promotion selector;

wherein said machine learning algorithm is configured to modify said key words and phrases associated with one or more of said set of one or more tags.