Identifying Negative Keywords Associated with Advertisements

• US 20120246016A1
• Filed: 06/11/2012
• Published: 09/27/2012
• Est. Priority Date: 04/17/2007
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: HUBINETTE, Carl F.
• Granted Patent: US 8,572,115 B2
• Time in Patent Office: Days
• US Class Current: 705/14.73
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising

Abstract

REVISE this to be consistent with the new claim 1 . . . . The present disclosure includes a system and method for identifying negative keywords associated with advertisements. In some implementations, a method for advertising includes identifying one or more keywords associated with an advertisement. The keywords used to embed the advertisement in web pages such as those including search results. By identifying, from the plurality of search criterion, a subset of search criterion as being off-topic to the advertisement item, one or more negative keywords may be derived from the subset of off-topic search criterion in order to limit the embedding of non-relevant advertisements in web page content.

Claims

1. A computer-implemented method comprising:

receiving, from a first data storage device, data encoding an advertisement item and a log comprising a plurality of previously submitted search criteria,

wherein each search criterion comprises one or more terms, and

wherein when the search criterion was evaluated, the advertisement item was presented in a corresponding search result webpage;

analyzing the plurality of previously submitted search criteria to identify one or more negative keywords including;

identifying, from the plurality of previously submitted search criteria, a subset of search criteria as being off-topic to the advertisement item, and

deriving the one or more negative keywords from the subset of off-topic search criteria;

generating one or more files including the one or more negative keywords, the one or more negative keywords being used to prevent the advertisement item from being presented in a search result corresponding to a later submitted search criterion containing at least one of the identified negative keywords; and

storing the one or more files on the first data storage device or a second data storage device.

2. The method of claim 1, further comprising:

selecting the negative keywords to increase a number of search criteria identified as being off-topic to the advertisement item.

3. The method of claim 2, wherein selecting the negative keywords to increase the number of search criteria identified as being off-topic to the advertisement item maximizes the number of search criteria identified as being off-topic to the advertisement item.

4. The method of claim 1, further comprising:

selecting the negative keywords to decrease the number of search criteria not identified as being off-topic to the advertisement item.

5. The method of claim 1, wherein selecting the negative keywords to decrease the number of search criteria not identified as off-topic to the advertisement item minimizes the number of search criteria not identified as being off-topic to the advertisement item.

6. The method of claim 1, wherein identifying the subset of search criteria as being off-topic to the advertisement item comprises:

associating, with each of the plurality of search criteria, a relevance score, and

if the relevance score is under a threshold relevance value, assigning the associated search criterion as being off-topic.

7. The method of claim 6, wherein the relevance score includes an average logarithmic odds that a term of the associated search criterion occurs in a set of keyword clusters associated with a topic of the advertisement item.

8. The method of claim 1, wherein deriving the one or more negative keywords from the subset of off-topic search criteria comprises:

identifying subsets of words in each of the off-topic search criteria;

associating each subset of words with an importance metric; and

summing the associated importance metric for each occurrence of the subset of words in the off-topic search criteria.

9. The method of claim 8, wherein the importance metric is determined by multiplying a confidence number by an amount of impressions associated with a set of keyword clusters that in turn are associated with the advertisement item, the confidence number representing a confidence that the identified off-topic search criterion is not relevant to a set of keyword clusters associated with a topic of the advertisement item.

10. The method of claim 8, wherein deriving the one or more negative keywords from the subset of off-topic search criteria further comprises:

when the summed importance metric is above a threshold importance value, assigning the subset of words as negative keywords.

11. The method of claim 8, deriving the one or more negative keywords from the subset of off-topic search criteria further comprises:

assigning the subset of words with the highest summed importance metric as negative keywords.

12. A non-transitory computer-readable medium having instructions stored thereon, which, when executed by at least one processor, cause the processor to perform operations comprising:

receiving, from a first data storage device, data encoding an advertisement item and a log comprising a plurality of previously submitted search criteria,

wherein each search criterion comprises one or more terms, and

wherein when the search criterion was evaluated, the advertisement item was presented in a corresponding search result webpage;

analyzing the plurality of previously submitted search criteria to identify one or more negative keywords including;

identifying, from the plurality of previously submitted search criteria, a subset of search criteria as being off-topic to the advertisement item, and

deriving the one or more negative keywords from the subset of off-topic search criteria;

generating one or more files including the one or more negative keywords, the one or more negative keywords being used to prevent the advertisement item from being presented in a search result corresponding to a later submitted search criterion containing at least one of the identified negative keywords; and

storing the one or more files on the first data storage device or a second data storage device.

13. The computer-readable medium of claim 12, wherein the operations further comprise:

selecting the negative keywords to increase a number of search criteria identified as being off-topic to the advertisement item.

14. The computer-readable medium of claim 13, wherein selecting the negative keywords to increase the number of search criteria identified as being off-topic to the advertisement item maximizes the number of search criteria identified as being off-topic to the advertisement item.

15. The computer-readable medium of claim 12, wherein the operations further comprise:

selecting the negative keywords to decrease the number of search criteria not identified as being off-topic to the advertisement item.

16. The computer-readable medium of claim 15, wherein selecting the negative keywords to decrease the number of search criteria not identified as being off-topic to the advertisement item minimizes the number of search criteria not identified as being off-topic to the advertisement item.

17. The computer-readable medium of claim 16, wherein identifying the subset of search criteria as being off-topic to the advertisement item comprises:

associating, with each of the plurality of search criteria, a relevance score, and

if the relevance score is under a threshold relevance value, assigning the associated search criterion as being off-topic.

18. The computer-readable medium of claim 17, the relevance score includes an average logarithmic odds that a term of the associated search criterion occurs in a set of keyword clusters associated with a topic of the advertisement item.

19. The computer-readable medium of claim 12, wherein deriving the one or more negative keywords from the subset of off-topic search criteria comprises:

identifying subsets of words in each of the off-topic search criteria;

associating each subset of words with an importance metric; and

summing the associated importance metric for each occurrence of the subset of words in the off-topic search criteria.

20. The computer-readable medium of claim 19, wherein the importance metric is determined by multiplying a confidence number by an amount of impressions associated with a set of keyword clusters that in turn are associated with the advertisement item, the confidence number representing a confidence that the off-topic search criterion is not relevant to a set of keyword clusters associated with a topic of the advertisement item.

21. The computer-readable medium of claim 19, wherein deriving the one or more negative keywords from the subset of off-topic search criteria further comprises:

when the summed importance metric is above a threshold importance value, assigning the subset of words as negative keywords.

22. The computer-readable medium of claim 19, wherein deriving the one or more negative keywords from the subset of off-topic search criteria further comprises:

assigning the subset of words with the highest summed importance metric as negative keywords.

23. A server for identifying negative words, comprising one or more processors, a memory coupled to the one or more processors, and a network interface, the one or more processors configured to:

receive data encoding an advertisement item and a log comprising a plurality of previously submitted search criteria,

wherein each search criterion comprises one or more terms, and

wherein when the search criterion was evaluated, the advertisement item was presented in a corresponding search result webpage;

evaluate the plurality of previously submitted search criteria to identify one or more negative keywords by;

identifying, from the plurality of previously submitted search criteria, a subset of search criteria as being off-topic to the advertisement item, and

deriving the one or more negative keywords from the subset of off-topic search criteria;

generate one or more files including the one or more negative keywords, the one or more negative keywords prevent the advertisement item from being presented in a search result corresponding to a later submitted search criterion containing at least one of the identified negative keywords; and

store the one or more files on computer-readable media.

24. The server of claim 23, wherein the one or more processors are further configured to select the negative keywords to increase the number of search criteria identified as being off-topic to the advertisement item.

25. The server of claim 24, wherein the one or more processors are further configured to select the negative keywords to further maximize the number of search criteria identified as being off-topic to the advertisement item.

26. The server of claim 23, wherein the one or more processors are further configured to select the negative keywords to decrease the number of search criteria not identified as being off-topic to the advertisement item.

27. The server of claim 26, wherein the one or more processors are further configured to select the negative keywords to further minimize the number of search criteria not identified as being off-topic to the advertisement item.

28. The server of claim 23, wherein the one or more processors are further configured to identify, from the plurality of search criteria, the subset of search criteria as being off-topic to the advertisement item by:

associating, with each of the plurality of search criteria, a relevance score, and

if the relevance score is under a threshold relevance value, assigning the associated search criterion as being off-topic.

29. The server of claim 28, wherein the relevance score include an average logarithmic odds that a term of the associated search criterion occurs in a set of keyword clusters associated with a topic of the advertisement item.

30. The server of claim 23, wherein the one or more processors are further configured to derive the one or more negative keywords from the subset of off-topic search criteria by:

identifying subsets of words in each of the off-topic search criteria;

associating each subset of words with an importance metric; and

summing the associated importance metric for each occurrence of the subset of words in the off-topic search criteria.

31. The server of claim 30, wherein the importance metric is determined by multiplying a confidence number by an amount of impressions associated with a set of keyword clusters that in turn are associated with the advertisement item, the confidence number representing a confidence that the off-topic search criterion is not relevant to a set of keyword clusters associated with a topic of the advertisement item.

32. The server of claim 30, wherein the one or more processors are further configured to derive the one or more negative keywords from the subset of off-topic search criteria by:

if the summed importance metric is above a threshold importance value, assigning the subset of words as negative keywords.

33. The server of claim 30, wherein the one or more processors are further configured to derive the one or more negative keywords from the subset of off-topic search criteria by:

assigning the subset of words with the highest summed importance metric as negative keywords.