Method and system for targeted advertising based on topical memes

• US 8,510,164 B2
• Filed: 04/23/2012
• Issued: 08/13/2013
• Est. Priority Date: 01/24/2008
• Status: Active Grant

• Current Assignee: Salesforce.com, Inc.
• Original Assignee: Salesforce.com, Inc.
• Inventors: Newton, Christopher Daniel, Lebrun, Marcel Albert, Ramsey, Christopher Bennett
• Primary Examiner(s): Hamilton, Matthew L
• Application Number:
  US13/453,802
• Publication Number:
  US 20120209702A1
• Time in Patent Office: 477 Days
• Field of Search: 705/14
• US Class Current: 705/14.4
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0224 based on user history

Abstract

A targeted advertising system and method based on memes contained in content sources are disclosed. Content matching keywords-defining topics are identified from content sources and are further processed to extract the memes. Ad networks servicing the content are also identified and their reach for each meme determined. The system and method extract also viral dynamics of the content associated to a meme and use the aggregation of the viral dynamics as a measure of engagement level for the meme. The system and method allow a Marketer to select a meme based on the engagement level and to run an ad campaign against the meme. The advertisements are delivered through an Ad network and inserted at the meme page level when the content hosting the meme is accessed, the Ad network being selected based on its reach.

Claims

1. A computer implemented method of targeting advertisement, comprising:

receiving, by a computer-based system, a topic selected by a user;

retrieving topic-matching content matching the selected topic;

extracting, by the computer-based system, a plurality of different topical memes from the topic-matching content, wherein each of the extracted topical memes is a point of discussion within the selected topic;

measuring an engagement level for each piece of the topic-matching content;

obtaining, based on the measuring, a respective viral dynamics metric for each said piece of the topic-matching content, wherein a viral dynamics metric indicates popularity of the respective piece of the topic-matching content;

accumulating, for each of the extracted topical memes, the viral dynamics metrics of the pieces of the topic-matching content;

obtaining, based on the accumulating, a respective aggregate viral dynamics value for each of the extracted topical memes;

determining a most active meme from the extracted topical memes, based on the aggregate viral dynamics value for each of the extracted topical memes;

running an advertisement campaign against the most active meme, wherein said advertisement campaign is targeted to web pages containing the most active meme;

selecting an advertising network servicing a number of content sources hosting the web pages containing the most active meme;

selecting an advertisement assigned to the most active meme; and

delivering the selected advertisement to the number of content sources, through the selected advertising network.

2. The method of claim 1, wherein each of the extracted topical memes is extracted by applying a feature extraction algorithm to the topic-matching content.

3. The method of claim 1, further comprising storing the aggregate viral dynamics value for each of the extracted topical memes in time series.

4. The method of claim 3, further comprising terminating the advertisement campaign when said aggregate viral dynamics value of the most active meme falls below a threshold.

5. The method of claim 1, wherein the advertising network is selected to maximize a number of pages containing the most active meme that can be reached by the advertisement campaign.

6. A system for performing meme-based targeted advertising, comprising:

a computer, having a processor and a computer readable storage medium storing computer readable instructions for execution by the processor, wherein the instructions, when executed by the processor, cause the computer to perform a method comprising;

retrieving topic-matching content matching a selected topic;

extracting a plurality of topical memes from the topic-matching content, wherein each of the extracted topical memes is a point of discussion within the selected topic;

measuring

an engagement level for each piece of the topic-matching content;

obtaining, based on the measuring, a respective viral dynamics metric for each said piece of the topic-matching content, wherein a viral dynamics metric indicates popularity of the respective piece of the topic-matching content;

accumulating, for each of the extracted topical memes, the viral dynamics metrics of the pieces of the topic-matching content;

obtaining, based on the accumulating, a respective aggregate viral dynamics value for each of the extracted topical memes;

determining a most active meme from the extracted topical memes, based on the aggregate viral dynamics value for each of the extracted topical memes;

running an advertisement campaign against said most active meme, wherein said advertisement campaign is targeted to web pages containing the most active meme;

selecting an advertising network servicing a number of content sources hosting the web pages containing the most active meme;

selecting an advertisement assigned to the most active meme; and

delivering the selected advertisement to the number of content sources, through the selected advertising network.

7. A non-transitory computer-readable medium comprising computer-executable instructions stored thereon, which, when executed by a computer, perform the steps comprising:

receiving a topic selected by a user;

retrieving topic-matching content matching the selected topic;

extracting a plurality of different topical memes from the topic-matching content, wherein each of the extracted topical memes is a point of discussion within the selected topic;

measuring an engagement level for each piece of the topic-matching content;

obtaining, based on the measuring, a respective viral dynamics metric for each piece of the topic-matching content, wherein a viral dynamics metric indicates popularity of the respective piece of the topic-matching content;

accumulating, for each of the extracted topical memes, the viral dynamics metrics of the pieces of the topic-matching content;

obtaining, based on the accumulating, a respective aggregate viral dynamics value for each of the extracted topical memes;

determining a most active meme from the extracted topical memes, based on the aggregate viral dynamics value for each of the extracted topical memes;

running an advertisement campaign against the most active meme, wherein said advertisement campaign is targeted to web pages containing the most active meme;

selecting an advertising network servicing a number of content sources hosting the web pages containing the most active meme;

selecting an advertisement assigned to the most active meme; and

delivering the selected advertisement to the number of content sources, through the selected advertising network.