System and method for automated mapping of keywords and key phrases to documents

• US 9,946,788 B2
• Filed: 07/22/2003
• Issued: 04/17/2018
• Est. Priority Date: 07/23/2002
• Status: Active Grant

• Current Assignee: Ouigo Tehchnologies Inc., Yahoo Assets LLC
• Original Assignee: Oath Inc. (Verizon Communications Inc.)
• Inventors: Itzhak, Ilan, Galai, Yaron, Itzhak, Oded
• Primary Examiner(s): Park, Grace
• Application Number:
  US10/485,108
• Publication Number:
  US 20040181525A1
• Time in Patent Office: 5,383 Days
• Field of Search: 707 3- 6, 707728, 704 9, 706 8
• CPC Class Codes: G06F 16/3347 using vector based model
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques

Abstract

A method for automated mapping of key terms to documents. Preferably a feature vector is generated for the key terms. Preferably such feature vectors are automatically generated by analyzing a corpus, but may optionally be generated manually, or using a combination of automated and manual processes. Next, preferably such feature vectors are weighted. Such weighting may optionally be performed manually, but more preferably is performed automatically. Next, a feature vector is optionally and preferably generated for the document, which preferably includes words and phrases that were extracted from the document but may optionally include words and phrases that do not appear in the document, such as synonyms and related. Each element in the document feature vector is preferably weighted. The feature vectors of the key terms and the feature vectors of the documents are compared, in order to produce relevancy scores, which are used to produce mapping between documents and key terms. This mapping may optionally be used for a wide variety of applications, such as for targeted advertising for example.

Claims

1. A system for selecting an advertisement for display with a web page requested by a user through a web browser, the system comprising:

at least one server, in communication with said web browser, configured to;

receive, as input, a plurality of documents and a plurality of key terms;

analyze content of the plurality of documents;

automatically generate a key term feature vector for each key term based on the content of the plurality of documents, each key term feature vector comprising elements comprising a plurality of words or phrases that are related to the corresponding key term;

analyze at least one of a Universal Resource Locator (“

URL”

) of said web page and a content of said web page;

automatically generate a document feature vector based on the URL or the content of the web page, the document feature vector comprising elements comprising a plurality of words or phrases that are contained in the URL or content of the web page;

compare the key term feature vector and the document feature vector;

generate a relevance factor based on the similarity of the elements of the key term feature vector and the document feature vector; and

provide a relevant advertisement for display by said web browser with said web page according to the determined relevance factor.

2. The system of claim 1, wherein said at least one server is configured to provide a plurality of advertisements.

3. The system of claim 1, wherein the at least one server is configured to provide the advertisement for display within the web page.

4. The system of claim 1, wherein the at least one server is configured to provide the advertisement for display on top of the web page.

5. The system of claim 1, wherein the at least one server is configured to provide the advertisement for display under the web page.

6. The system of claim 1, wherein the at least one server is configured to:

receive said URL for said web page from said web browser; and

parse at least one of said URL and said content of said web page to generate said document feature vector of said URL or of said content of said web page.

7. The system of claim 6, wherein the at least one server is configured to generate said document feature vector of said URL or of said content of said web page in real time in response to said receipt of said URL from said web browser.

8. The system of claim 6, wherein the at least one server is configured to:

generate said document feature vector of said URL or of said content of said web page prior to said receipt of said URL from said web browser;

store said document feature vector of said URL or said content of said content of said web page in a database; and

retrieve the document feature vector from said database in real time in response to said receipt of said URL from said web browser.

9. The system of claim 8, further comprising a module that monitors the web page on an ongoing basis to alert when at least one of the web page structure, URL structure, and content on the web page has changed.

10. The system of claim 6, wherein said at least one server is configured to generate said document feature vector of said content of said web page by accessing said web page and determining a frequency of words or phrases in said content of said web page.

11. The system of claim 6, wherein said at least one server is configured to generate said document feature vector of said content of said web page by accessing said web page and using machine learning to extract information from said content of said web page.

12. The system of claim 6, wherein said document feature vector of said URL or of said content of said web page is a weighted feature vector.

13. The system of claim 1, wherein the at least one server is configured to:

receive a plurality of documents from said web browser; and

parse the plurality of documents to generate the key term feature vector of the plurality of documents.

14. The system of claim 13, wherein said at least one server is configured to generate the key term feature vector of the plurality of documents by accessing the plurality of documents and determining a frequency of words or phrases in the contents of the plurality of documents.

15. The system of claim 13, wherein said at least one server is configured to generate the key term feature vector of the plurality of documents by accessing the plurality of documents and using machine learning to extract information from the content of the plurality of documents.

16. The system of claim 13, wherein the key term feature vector of the plurality of documents is a weighted feature vector.

17. A method for selecting advertising for display with a web page requested by a user comprising:

receiving, as input, a plurality of documents and a plurality of key terms;

analyzing content of the plurality of documents;

automatically generating a key term feature vector for each key term based on the content of the plurality of documents, each key term feature vector comprising elements comprising a plurality of words or phrases that are related to the corresponding key term;

analyzing at least one of a Universal Resource Locator (“

URL”

) of said web page and a content of said web page;

automatically generating a document feature vector based on the URL or the content of the web page, the document feature vector comprising elements comprising a plurality of words or phrases that are contained in the URL or content of the web page;

comparing the key term feature vector and the document feature vector;

generating a relevance factor based on the similarity of the elements of the key term feature vector and the document feature vector; and

providing a relevant advertisement for display by said web browser with said web page according to the determined relevance factor.

18. The method of claim 17, further providing a plurality of advertisements.

19. The method of claim 17, wherein said providing an advertisement for display comprises providing the advertisement for display within the web page.

20. The method of claim 17, wherein said providing an advertisement for display comprises providing the advertisement for display on top of the web page.

21. The method of claim 17, wherein said providing an advertisement for display comprises providing the advertisement for display under the web page.

22. The method of claim 17, further comprising:

receiving said URL for said web page;

parsing at least one of said URL and said content of said web page; and

based on said parsing, generating said document feature vector of said URL or of said content of said web page.

23. The method of claim 22, wherein said generating said document feature vector comprises generating said document feature vector of said URL or of said content of said web page in real time in response to said receiving said URL for said web page.

24. The method of claim 22, wherein said generating said document feature vector comprises generating said document feature vector of said URL or of said content of said web page prior to said receiving said URL for said web page, wherein said method further comprises:

storing said document feature vector of said URL or of said content of said web page and

retrieving the document feature vector from said database in real time in response to said receiving said URL for said web page.

25. The method of claim 24, further comprising monitoring said web page on an ongoing basis to alert when at least one of the web page structure, URL structure, and said content of said web page has changed.

26. The method of claim 22, wherein said generating said document feature vector comprises generating said document feature vector of said content of said web page by accessing said web page and determining a frequency of words or phrases in said content of said web page.

27. The method of claim 22, wherein said generating said document feature vector comprising generating said document feature vector of said content of said web page by accessing said web page and using machine learning to extract information from said content of said web page.

28. The method of claim 22, wherein said generating said document feature vector comprises generating a weighted feature vector.

29. The method of claim 17, further comprising:

receiving a plurality of documents from said web browser; and

parsing the plurality of documents to generate the key term feature vector of the plurality of documents.

30. The method of claim 29, further comprising generating the key term feature vector of the plurality of documents by accessing the plurality of documents and determining a frequency of words or phrases in the contents of the plurality of documents.

31. The method of claim 29, further comprising generating the key term feature vector of the plurality of documents by accessing the plurality of documents and using machine learning to extract information from the content of the plurality of documents.

32. The method of claim 29, wherein the key term feature vector of the plurality of documents is a weighted feature vector.