Clickable terms for contextual advertising

• US 8,533,043 B2
• Filed: 03/31/2010
• Issued: 09/10/2013
• Est. Priority Date: 03/31/2010
• Status: Expired due to Fees

• Current Assignee: R2 Solutions LLC (Acacia Research Corporation)
• Original Assignee: Yahoo! Inc. (Apollo Global Management, Inc.)
• Inventors: Ratnaparkhi, Adwait, Bagherjeiran, Abraham, Hatch, Andrew O.
• Primary Examiner(s): Weiss, John G
• Assistant Examiner(s): Anderson, Scott C
• Application Number:
  US12/751,937
• Publication Number:
  US 20110246285A1
• Time in Patent Office: 1,259 Days
• Field of Search: 705/14.6, 705/14.42
• US Class Current: 705/14.43
• CPC Class Codes: G06Q 30/00 Commerce
• CPC Class Codes: G06Q 30/0243 Comparative campaigns

Abstract

An online advertising selects online advertisements for display on a network location taking into account a probability that a candidate online advertisement will receive a click on a particular website. The system may determine a network location identity of the network location and transform a set of advertisements into a set of ranked advertisements. The system may determine an advertisement rank of a first advertisement among the set of ranked advertisements. The system then may generate a click probability value. The click probability value may reflect a click probability of the first advertisement by dividing an exponent of a weighted sum of the network location identity and the advertisement rank by one plus the exponent of the weighted sum of the network location identity and the advertisement rank.

Claims

1. An online advertising method implemented in one or more computers, each computer comprising at least one processor and memory, to select advertisements for display on a webpage, comprising:

storing, in a computer, a website identity for a webpage, wherein the website identity comprises one or more terms derived from the uniform resource locators (URLs) of the website;

storing, in a computer, a set of candidate advertisements;

estimating, in a computer, a click probability for each candidate advertisement based on the website identity; and

selecting, in a computer, at least one candidate advertisement for displaying on the webpage based on a highest click probability associated with the candidate advertisement.

2. The online advertising method of claim 1, wherein estimating a click probability comprises estimating the click probability based on the degree to which terms from a candidate advertisement match the website identity.

3. The online advertising method of claim 1, wherein estimating a click probability comprises a click model that is based on a ranking of the set of candidate advertisements and the website identity.

4. The online advertising method of claim 3, wherein the click model employs the equation:

5. The online advertising method of claim 4, wherein the click model employs the equation:

6. The online advertising method of claim 3, wherein the click model is based on the ranking, the website identity, and a set of terms from a candidate advertisement.

7. The online advertising method of claim 6, wherein the click model provides a measure for a term from a candidate advertisement, wherein the measure may be used to estimate whether the term is more or less likely to be clicked upon than a second term.

8. The online advertising method of claim 6, wherein the click model employs the equation:

9. The online advertising method of claim 6, wherein the click model defines one separate model weight for every combination of website identity and term from the set of terms.

10. The online advertising method of claim 8, wherein, in the equation,

11. The online advertising method of claim 9, wherein the click model employs a term importance weight that reflects the relative importance of a term, and wherein the term importance weight is a function of a total number of unique terms, or a function of an expected number of unique terms that co-occur with a given term in advertisements that appear on a webpage.

12. The online advertising method of claim 7, wherein the click model employs a term importance weight that reflects the importance of the term relative to other terms within the set.

13. The online advertising method of claim 12, wherein the click model employs a term importance weight that reflects the relative importance of a term, and wherein the term importance weight is a function of a total number of unique terms, or a function of an expected number of unique terms that co-occur with a given term in advertisements that appear on a webpage.

14. The online advertising method of claim 8, wherein

15. The online advertising method of claim 7, wherein the set of terms from a candidate advertisement is limited to terms whose measure passes a threshold.

16. The online advertising method of claim 15, wherein the click model employs a term importance weight that reflects the relative importance of a term, and wherein the term importance weight is a function of a total number of unique terms, or a function of an expected number of unique terms that co-occur with a given term in advertisements that appear on a webpage.

17. A computer program product stored on a non-transitory computer-readable medium that when executed by a processor enables the processor to select advertisements for display on a webpage, comprising:

first computer program logic means for storing a website identity for a webpage, wherein the website identity comprises one or more terms derived from the URLs of the website;

second computer program logic means for storing a set of candidate advertisements;

third computer program logic means for estimating a click probability for each candidate advertisement based on the website identity; and

fourth computer program logic means for selecting at least one candidate advertisement for displaying on the webpage based on a highest click probability associated with the candidate advertisement.