Advertising keyword cross-selling

• US 7,788,131 B2
• Filed: 12/15/2005
• Issued: 08/31/2010
• Est. Priority Date: 12/15/2005
• Status: Expired due to Fees

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Chen, Zheng, Zhang, Benyu, Zeng, Hua-Jun, Najm, Tarek, Li, Li, Li, Ying, Nong, Shuzhen
• Primary Examiner(s): Myhre; James W
• Assistant Examiner(s): CHORNESKY, ADAM B
• Application Number:
  US11/300,918
• Publication Number:
  US 20070143176A1
• Time in Patent Office: 1,720 Days
• Field of Search: 705/14, 705/10, 705/14.49, 705/14.71
• US Class Current: 705/14.49
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0251 Targeted advertisements
• CPC Class Codes: G06Q 30/0275 Auctions

Abstract

Seed keywords are leveraged to provide expanded keywords that are then associated with relevant advertisers. Instances can also include locating potential advertisers based on the expanded keywords. Inverse lookup techniques are employed to determine which keywords are associated with an advertiser. Filtering can then be employed to eliminate inappropriate keywords for that advertiser. The keywords are then automatically revealed to the advertiser for consideration as relevant search terms for their advertisements. In this manner, revenue for a search engine and/or for an advertiser can be substantially enhanced through the automatic expansion of relevant search terms. Advertisers also benefit by having larger and more relevant search term selections automatically available to them, saving them both time and money.

Claims

1. A computer system, including one or more computer storage media having computer-executable components stored thereon, that facilitates advertising keyword cross-selling, the computer-executable components comprising:

a receiving component that receives search keywords from a keyword data source;

a keyword determination component that;

selects, as seed keywords, a subset of the received search keywords and

expands the seed keywords to determine other keywords relevant to the seed keywords, the seed keywords being selected based on at least one of commercial value, frequency of use, and a number of times the seed keywords have previously been bid on by advertisers;

a cross-sell pair determination component that pairs the expanded keywords with relevant advertisers to facilitate in cross-selling the expanded keywords to the relevant advertisers; and

a proactive cross-sell component that actively promotes the expanded keywords to the relevant advertisers by automatically contacting the relevant advertisers with suggestions of suitable keywords selected from the expanded keywords via an electronic medium.

2. The system of claim 1, wherein the keyword data source comprises a search query log.

3. The system of claim 1,

wherein the keyword determination component employs a funnel search process that analyzes similar interest between the seed keywords and the other keywords to facilitate in determining the expanded keywords,

wherein the funnel search process includes the step of forming search funnels that represent transitions utilized by a plurality of searchers from a first search keyword to at least one additional search keyword,

and wherein the size of each search funnel is determined by a number of the plurality of searchers that search with both the first search keyword and the at least one additional search keyword.

4. The system of claim 1, further comprising:

a cross-linking component that indexes the expanded keywords to the relevant advertisers to facilitate in the construction of cross-sell pairs.

5. The system of claim 1 further comprising:

a cross-selling filter component that filters inappropriate cross-sell pairs based on attitude and/or cause-effect relationships.

6. The system of claim 5, wherein the cross-selling filter component employs a machine learning classifier to facilitate in filtering the inappropriate cross-sell pairs.

7. An online advertising auction process that employs the system of claim 1.

8. An online advertising keyword service that employs the system of claim 1 to provide potential advertising keywords to a user.

14. A device employing the system of claim 1 comprising at least one selected from the group consisting of a computer, a server, or a handheld electronic device.

15. The system of claim 1, wherein the keyword data source comprises an advertisement data source.

16. The system of claim 1, wherein the automatic contacting of the advertisers with suggestions of suitable keywords is unsolicited.

17. The system of claim 1, wherein automatically contacting the advertisers with suggestions of suitable keywords is performed by providing the suggestions of suitable keywords in billing statements sent to the advertisers.

9. One or more computer storage media having computer-useable instructions embodied thereon for performing a method for facilitating advertising keyword cross-selling, the method comprising:

obtaining search keywords from a keyword data source;

determining a commercial value for at least a portion of the search keywords;

selecting as seed keywords a subset of the obtained search keywords that have commercial value over a designated threshold;

after selecting the seed keywords, expanding the seed keywords through one of search funnel expansion or search result extraction to determine sequentially and/or conceptually related keywords, wherein search result extraction comprises searching for the seed keywords in a search engine, obtaining the top web page titles and content resulting from the searching, and merging the top web page titles and content;

automatically discovering potentially commercially valuable keywords from the sequentially and/or conceptually related keywords;

constructing a keyword-to-advertiser inverse index table associating the potentially commercially valuable sequentially and/or conceptually related keywords identified after expanding the seed keywords with advertisers relevant to the keywords;

retrieving a list of the relevant advertisers by querying the keyword-to-advertiser inverse index table with the potentially commercially valuable sequentially and/or conceptually related keywords identified after expanding the seed keywords;

assigning a score to advertiser/keyword pairs retrieved from the querying;

determining a top N most relevant advertisers, wherein N is an integer from one to infinity;

filtering inappropriate cross-selling by analyzing attitude and/or cause-effect relationships; and

proactively cross-selling the potentially commercially valuable keywords to the relevant advertisers by automatically informing the relevant advertisers with the potentially commercially valuable keywords.

10. The media of claim 9, wherein the keyword data source comprises a search query log.

11. The media of claim 9, wherein the method further comprises:

discovering the relevant advertisers based on the potential commercially valuable keywords via a global network.

18. The media of claim 9, wherein the keyword data source comprises an advertisement data source.

12. One or more computer storage media having computer-useable instructions embodied thereon for performing a method that facilitates keyword expansion for relevant advertisers, the method comprising:

obtaining search keywords from a keyword data source;

selecting as seed keywords a subset of the obtained search keywords that have commercial value, are frequently used, and/or have been bid on by advertisers less than a specified number of times;

obtaining search page views generated for the seed keywords from a search data source;

determining a session to which the search page views belong;

extracting other search keywords from each of one or more sessions in order and forming search funnels;

counting a number of searchers for each other search keyword to determine sizes of the search funnels;

identifying relevant advertisers for the other search keywords; and

automatically suggesting one or more of the other search keywords to the relevant advertisers.

13. The media of claim 12, wherein the method further comprises:

consolidating the search funnels utilizing funnel clustering, noise filtering, and/or keyword-type categorization.