Recommending search terms using collaborative filtering and web spidering

• US 7,225,182 B2
• Filed: 12/11/2001
• Issued: 05/29/2007
• Est. Priority Date: 05/28/1999
• Status: Expired due to Term

• Current Assignee: R2 Solutions LLC (Acacia Research Corporation)
• Original Assignee: Overture Services Incorporated (Apollo Global Management, Inc.)
• Inventors: Davis, Darren J., Davies, Winton, Geddis, Donald F., Paine, Mark, Dukes-Schlossberg, Jon
• Primary Examiner(s): LEROUX, ETIENNE PIERRE
• Application Number:
  US10/020,712
• Publication Number:
  US 20030055816A1
• Time in Patent Office: 1,995 Days
• Field of Search: 707/1, 707/3, 707/4, 707/5, 707/7, 707/10, 707/104.1, 707/2, 345/854, 702/2, 705/51, 704/9, 709/226
• US Class Current: 707/709
• CPC Class Codes: G06F 16/3322 using system suggestions G06F16/3325 takes precedence
• CPC Class Codes: G06F 16/36 Creation of semantic tools, e.g. ontology or thesauri
• 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: Y10S 707/918 Location
• CPC Class Codes: Y10S 707/959 Network
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching

Abstract

In a pay-for-placement search system, the system makes search term recommendations to advertisers managing their accounts in one or more of two ways. A first technique involves looking for good search terms directly on an advertiser'"'"'s web site. A second technique involves comparing an advertiser to other, similar advertisers and recommending the search terms the other advertisers have chosen. The first technique is called spidering and the second technique is called collaborative filtering. In the preferred embodiment, the output of the spidering step is used as input to the collaborative filtering step. The final output of search terms from both steps is then interleaved in a natural way.

Claims

1. A method for recommending search terms in a computer network search apparatus for generating a result list of items representing a match with information entered by a user through an input device connected to the computer network, the search apparatus including a computer system operatively connected to the computer network and a plurality of items stored in a database, each item including information to be communicated to a user and having associated with it at least one search term, an information provider and a bid amount, the method comprising:

(a) obtaining a set of potential search terms for acceptance by a new information provider who is adding items to the database, including;

receiving from the new information provider a website uniform resource locator (URL); and

spidering the website associated with the website URL to obtain search terms for the set of potential search terms;

(b) computing correlations between the potential search terms for the new information provider and search terms of other information providers stored in the database;

(c) computing an estimated rating for the each potential search term for the new information provider;

(d) sorting the potential search terms according to the computed estimated ratings;

(e) presenting to the new information provider on an output device the sorted potential search terms;

(f) receiving from the new information provider at an input device an indication of accepted search terms;

(g) repeating (b) through (e) until a completion indication is received from the new information provider; and

(h) storing the accepted search terms in the database for the new information provider upon receipt of the completion indication.

2. The method of claim 1 wherein spidering the website comprises:

receiving data from pages of the website;

recording potential search terms from the data; and

determining a quality metric for each potential search term.

3. The method of claim 2 wherein computing an estimated rating comprises:

combining a rating based on the computed correlations and a rating based on the quality metric determined for each candidate search term.

4. The method of claim 2 further comprising:

sorting the candidate search terms according to the quality metric; and

adding to the set of potential search terms only candidate search terms having a quality metric exceeding a threshold.

5. The method of claim 1 wherein spidering comprises:

receiving data from one or more pages of the website; and

examining text from the one or more pages for candidate search terms.

6. The method of claim 5 wherein examining text comprises:

examining substantially all text from the one or more pages; and

examining meta tags from the one or more pages.

7. The method of claim 5 wherein receiving a website URL comprises:

receiving the advertiser'"'"'s URL as the web site URL.

8. The method of claim 5 wherein receiving a website URL comprises:

receiving the web site URL from the advertiser.

9. The method of claim 1 wherein computing correlations comprises:

assigning ratings to search terms; and

computing a correlation between the advertiser and one or more of the other advertisers using the assigned ratings of advertiser search terms.

10. The method of claim 9 wherein computing an estimated rating comprises:

predicting a likelihood that a search term will be relevant to the advertiser.

11. The method of claim 10 wherein predicting comprises:

determining a quality metric for potential search terms; and

predicting relevance of the potential search terms based on the quality metric.

12. The method of claim 1 wherein presenting the sorted potential search terms to the new information provider comprises sending the sorted potential search terms with a web page to the output device.

13. A computer network search engine apparatus which includes a database having stored therein a plurality of search listings, each search listing being associated with an information provider, at least one keyword, a money amount and a computer network location and a search engine to identify search listings having a keyword matching a keyword entered by a searcher, to order the identified listings using the money amounts for the respective identified listings, and to generate a result list including at least some of the ordered listings, the apparatus comprising:

an account management server including a processing system which is operative in conjunction with program code to recommend potential search terms to a new information provider adding search listings to the database;

collaborative filtering code operable in conjunction with the processing system to compute correlations between potential search terms for the new information provider and search terms of other information providers stored in the database and to compute an estimated rating for the each potential search term for the new information provider;

sorting code operable in conjunction with the processing system and configured to sort the potential search terms according to the computed estimated ratings;

spidering code overable in conjunction with the processing system to find initially accepted search terms in a web site by spidering the web site and to include the initially accepted search terms among the sorted potential search terms;

an output device configured to provide the sorted potential search terms to the new information provider for review; and

an input device configured to receive from the new information provider an indication of accepted search terms, the accepted search terms being stored in the database in association with the new information provider upon receipt of the indication from the new information provider.

14. The computer network search engine apparatus of claim 13 wherein the spidering code is configured to spider a web site of the new information provider.

15. The computer network search apparatus of claim 14 wherein the spidering code is configured to spider a web site specified by the new information provider.

16. The computer network search engine apparatus of claim 13 wherein the spidering code is configured to retrieve pages from the web site of the new information provider, record terms contained in the retrieved pages and score the terms according to a quality metric.

17. The computer network search engine apparatus of claim 16 wherein the spidering code is configured to include terms scoring above a threshold score among the sorted potential search terms.

18. A method for making search term recommendations to an advertiser in a pay for placement market system in which search listings of advertisers may be searched by users entering search terms, the method comprising:

receiving from the advertiser a website uniform resource locator (URL);

spidering the website associated with the website URL to obtain an initial list of search terms to form a set of potential search terms for the advertiser;

computing correlations between the set of potential search terms for the advertiser and search terms of other advertisers stored in a database of the pay for placement market system;

computing an estimated rating for each potential search term for the advertiser;

sorting the potential search terms according to the estimated ratings;

providing the sorted potential search terms to the advertiser;

receiving from the advertiser the advertiser'"'"'s indication of accepted search terms; and

storing the accepted search terms in the database for searching by the users.

19. The method of claim 18 further comprising:

repeating the acts of computing correlations, computing an estimated rating, sorting and providing the potential search terms and receiving an indication of accepted search terms until the advertiser indicated the process is complete.