Method and system for identifying keywords for use in placing keyword-targeted advertisements

• US 10,402,431 B2
• Filed: 11/04/2016
• Issued: 09/03/2019
• Est. Priority Date: 08/09/2004
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Scholl, Nathaniel B., DeNeui, Alexander W.
• Primary Examiner(s): Trujillo, James
• Assistant Examiner(s): Khoshnoodi, Fariborz
• Application Number:
  US15/344,200
• Publication Number:
  US 20170103122A1
• Time in Patent Office: Days
• Field of Search: 707769, 707770
• CPC Class Codes: G06F 16/313 Selection or weighting of terms for indexing
• CPC Class Codes: G06F 16/3325 Reformulation based on results of preceding query
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0256 User search

Abstract

A method and system for identifying search terms for placing advertisements along with search results is provided. The advertisement system selects a description of an item that is to be advertised. The advertisement system then retrieves documents that match the selected description. The advertisement system generates a score for each word of the retrieved documents that indicates relatedness of the word to the item to be advertised. After generating the scores for the words, the advertisement system identifies phrases of the words within the documents that are related to the item. The advertisement system then generates search terms for the item to be advertised from the identified phrases. The advertisement system submits the search terms and an advertisement to a search engines service for placement of a paid-for advertisement for the item.

Claims

1. A computer system for identifying phrases related to an item, comprising:

a data store storing information from one or more documents related to the item;

at least one processor operable to access the information from the data store, and the at least one processor operable to cause the computer system to;

determine a result set including a plurality of documents from a corpus of documents, wherein each document in the plurality of documents of the result set is related to the item;

determine a first frequency of at least one word in the plurality of documents of the result set, wherein the first frequency corresponds to an average number of times that the at least one word appears in each document of the result set;

determine a second frequency of the at least one word in the plurality of documents of the corpus of documents, wherein the second frequency corresponds to an average number of times the at least one word appears in each document of the corpus of documents;

determine a frequency score that is based on a difference between the corresponding first frequency and the corresponding second frequency for the at least one word;

identify a set of highly related words to the item based on a threshold number of the at least one word having a highest frequency score;

select at least one anchor word from the set of highly related words; and

identify at least one phrase in the plurality of documents in the result set that contains the selected at least one anchor word by searching the plurality of documents in the search result set for the at least one anchor word and identifying at least one word proximal to the at least one anchor word within the plurality of documents.

2. The computer system of claim 1, wherein the at least one processor is operable to cause the computing system to identify the at least one phrase by, at least:

identifying a first phrase that contains the selected at least one anchor word;

determining whether extending the first phrase by one or more words proximal to the first phrase resulting in an extended first phrase is more relevant to the item than the first phrase; and

identifying the extended first phrase as the at least one phrase.

3. The computing system of claim 2, wherein determining whether extending the first phrase includes identifying, based at least in part on the determined frequency score of the one or more words proximal to the first phrase, the one or more words proximal to the first phrase.

4. The computing system of claim 1, wherein the frequency score indicates a level of relevance of the at least one word to the item.

5. The computing system of claim 1, wherein the at least one processor is operable to cause the computing system to select the at least one anchor word so as to exclude noise words.

6. The computing system of claim 1, wherein the at least one processor is operable to cause the computing system to:

generate one or more search terms for the item based on the at least one phrase; and

submit a request to place a content item for the item to a search engine service, the request including the one or more search terms.

7. A computing-implemented method for identifying phrases related to an item, the method comprising:

determining a result set including a plurality of documents from a corpus of documents, wherein each document in the plurality of documents of the result set is related to the item;

determining a first frequency of at least one word in the plurality of documents of the result set, wherein the first frequency corresponds to an average number of times that the at least one word appears in each document of the result set;

determining a second frequency of the at least one word in the plurality of documents of the corpus of documents, wherein the second frequency corresponds to an average number of times the at least one word appears in each document of the corpus of documents;

determining a frequency score that is based on a difference between the corresponding first frequency and the corresponding second frequency for the at least one word;

identifying a set of highly related words to the item based on a threshold number of the at least one word having a highest frequency score;

selecting at least one anchor word from the set of highly related words; and

identifying at least one phrase in the plurality of documents in the result set that contains the selected at least one anchor word by searching the plurality of documents in the search result set for the at least one anchor word and identifying at least one word proximal to the at least one anchor word within the plurality of documents.

8. The computer-implemented method of claim 7, wherein identifying at least one phrase in the plurality of documents in the result set comprises:

identifying a first phrase that contains the selected at least one anchor word;

determining whether extending the first phrase by one or more words proximal to the first phrase resulting in an extended first phrase is more relevant to the item than the first phrase; and

identifying the extended first phrase as the at least one phrase.

9. The computer-implemented method of claim 8, wherein determining whether extending the first phrase includes identifying, based at least in part on the determined frequency score for the one or more words proximal to the first phrase, the one or more words proximal to the first phrase.

10. The computer-implemented method of claim 7, further comprising:

generating one or more search terms for the item based on the at least one phrase; and

submitting a request to place a content item for the item to a search engine service, the request including the one or more search terms.

11. The computer-implemented method of claim 7, wherein the frequency score indicates a level of relevance of the at least one word to the item.

12. The computer-implemented method of claim 7, further comprising selecting the at least one anchor word so as to exclude words that have a second frequency over a second frequency threshold.

13. The computer-implemented method of claim 7, wherein the at least one processor is operable to cause the computing system to select the at least one anchor word so as to exclude noise words.

14. A computer-readable storage medium having stored thereon instructions for causing one or more computing systems to perform a method of identifying phrases related to an item, the method comprising:

determining a result set including a plurality of documents from a corpus of documents, wherein each document in the plurality of documents of the result set is related to the item;

determining a first frequency of at least one word in the plurality of documents of the result set, wherein the first frequency corresponds to an average number of times that the at least one word appears in each document of the result set;

determining a second frequency of the at least one word in the plurality of documents of the corpus of documents, wherein the second frequency corresponds to an average number of times the at least one word appears in each document of the corpus of documents;

determining a frequency score that is based on a difference between the corresponding first frequency and the corresponding second frequency for the at least one word;

identifying a set of highly related words to the item based on a threshold number of the at least one word having a highest frequency score;

selecting at least one anchor word from the set of highly related words; and

identifying at least one phrase in the plurality of documents in the result set that contains the selected at least one anchor word by searching the plurality of documents in the search result set for the at least one anchor word and identifying at least one word proximal to the at least one anchor word within the plurality of documents.

15. The computer-readable storage medium of claim 14, wherein identifying at least one phrase in the plurality of documents in the result set comprises:

identifying a first phrase that contains the selected at least one anchor word;

determining whether extending the first phrase by one or more words proximal to the first phrase resulting in an extended first phrase is more relevant to the item than the first phrase; and

identifying the extended first phrase as the at least one phrase.

16. The computer-readable storage medium of claim 15, wherein determining whether extending the first phrase includes identifying, based at least in part on the determined frequency score of the one or more words proximal to the first phrase, the one or more words proximal to the first phrase.

17. The computer-readable storage medium of claim 16, wherein the determining whether extending the first phrase includes excluding one or more words similar to the at least one anchor word in the first phrase.

18. The computer-readable storage medium of claim 14, further comprising selecting the at least one anchor word so as to exclude noise words.

19. The computer-readable storage medium of claim 14, wherein the frequency score indicates a level of relevance of the at least one word to the item.

20. The computer-readable storage medium of claim 14, wherein the method further comprises:

generating one or more search terms for the item based on the at least one phrase; and

submitting a request to place a content item for the item to a search engine service, the request including the one or more search terms.