Determining Phrases Related to Other Phrases

• US 20100179801A1
• Filed: 01/13/2009
• Published: 07/15/2010
• Est. Priority Date: 01/13/2009
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Vaidya, Vinay, Huynh, Steve, Jesensky, James, Oates, Isaac
• Granted Patent: US 8,768,852 B2
• Time in Patent Office: Days
• US Class Current: 704/1
• CPC Class Codes: G06F 40/56 Natural language generation

Abstract

Techniques for generating and providing phrases are described herein. These techniques may include analyzing one or more sources to generate a first corpus of phrases, each of the phrases for use as an identifier and/or for association with a user. Once a first corpus of phrases has been generated, these phrases may be filtered to define a second corpus of phrases. Phrases of this second corpus may then be suggested to one or more users. In some instances, the phrases suggested to a particular user are personalized to the user based on information previously known about the user or based on information provided by the user.

Claims

1. A method implemented at least in part by a computing device, the method comprising:

creating an index that maps each of multiple keywords to one or more phrases that each comprise at least one word;

determining one or more statistically improbable phrases associated with a user, each of the one or more statistically improbable phrases comprising at least two contiguous words that occur in a corpus of text more than a threshold number of times;

determining the words that compose the one or more statistically improbable phrases that are associated with the user;

inputting, into the index as keywords, the determined words that compose the one or more statistically improbable phrases to determine phrases that are associated with the determined words that compose the one or more statistically improbable phrases; and

outputting the determined phrases for suggestion to the user.

2. One or more computer-readable media storing computer-executable instructions that, when executed on one or more processors, perform acts comprising:

determining one or phrases that share a common characteristic, each of the phrases comprising one or more grammatically-correct words;

determining words that compose the one or phrases that share the common characteristic;

determining phrases that are associated with at least one of the determined words; and

providing at least a portion of the determined phrases for selection by a user.

3. One or more computer-readable media as recited in claim 2, wherein the determining of the phrases that are associated with at least one of the determined words comprises analyzing an index that maps each of multiple words to one or more phrases.

4. One or more computer-readable media as recited in claim 2, further storing computer-executable instructions that, when executed on one or more processors, perform acts comprising:

receiving a selection of a phrase from the user; and

responsive to the receiving of the selection, associating the selected phrase with the user.

5. One or more computer-readable media storing computer-executable instructions that, when executed on one or more processors, perform acts comprising:

determining a statistically improbable phrase that appears in a source;

determining words that compose the statistically improbable phrase;

determining phrases that are associated with at least one of the determined words; and

providing at least a portion of the determined phrases for selection by a user.

6. One or more computer-readable media as recited in claim 5, wherein the source comprises a book, a magazine, online content, audio content or video content.

7. One or more computer-readable media as recited in claim 5, wherein the determining of the phrases comprising analyzing an index that maps each of multiple words to one or more phrases.

8. One or more computer-readable media as recited in claim 5, further storing computer-executable instructions that, when executed on one or more processors, perform acts comprising:

receiving a selection of a phrase from the user; and

responsive to the receiving of the selection, associating the selected phrase with the user.

9. One or more computer-readable media as recited in claim 5, further storing computer-executable instructions that, when executed on one or more processors, perform acts comprising:

receiving a selection of a phrase from the user; and

responsive to the receiving of the selection, associating the selected phrase with the user and with a payment instrument of the user.

10. One or more computer-readable media as recited in claim 5, further storing computer-executable instructions that, when executed on one or more processors, perform acts comprising:

determining that the source includes multiple statistically improbable phrases;

determining words that compose the multiple statistically improbable phrases; and

determining phrases that are associated with at least one of the determined words that compose the multiple statistically improbable phrases.

11. One or more computer-readable media as recited in claim 10, wherein the determined multiple statistically improbable phrases comprise each statistically improbable phrase that appears in the source.

12. A method implemented at least in part by a computing device, the method comprising:

determining, for each phrase of a first set of phrases, words that compose the respective phrase;

determining a second set of phrases, each phrase of the second set of phrases being associated with at least one of the determined words that compose the phrases of the first set of phrases;

suggesting at least a portion of the determined second set of phrases to a user; and

responsive to receiving a selection of a phrase from the user, associating the selected phrase with the user.

13. A method as recited in claim 12, wherein the first set of phrases comprise phrases that share a common characteristic.

14. A method as recited in claim 12, wherein at least a portion of the first set of phrases comprise statistically improbable phrases.

15. A method as recited in claim 12, wherein the determining of the second set of phrases comprises:

inputting, into an index that maps keywords to phrases, the determined one or more words of the first set of phrases as keywords to the index; and

determining phrases that are associated with one or more of the inputted keywords.

16. A method as recited in claim 12, further comprising determining a degree of relation between the first set of phrases and each phrase of the second set phrases that are determined to be associated with the first set of phrases.

17. A method as recited in claim 12, further comprising determining a degree of relation between the first set of phrases and each phrase of the second set of phrases that are determined to be associated with first set of phrases, and wherein the suggesting of the portion of determined phrases comprises suggesting phrases having a relatively higher degree of relation to the first set of phrases.

18. A system comprising:

one or more processors; and

the one or more computer-readable media storing the computer-executable instructions as recited in claim 12.

19. One or more computer-readable media storing computer-executable instructions that, when executed on one or more processors, perform acts comprising:

determining that a phrase is associated with a user;

determining words that compose the phrase;

analyzing an index with use of the determined words to determine one or more other phrases; and

providing at least a portion of the determined one or more other phrases for selection by the user.

20. One or more computer-readable media as recited in claim 19, wherein the phrase comprises a statistically improbable phrase, and wherein the determining of the statistically improbable phrase that is associated with the user comprises:

determining a keyword that is associated with the user; and

determining a statistically improbable phrase that is associated with the keyword.

21. One or more computer-readable media as recited in claim 20, wherein the keyword is manually inputted by the user.

22. One or more computer-readable media as recited in claim 20, wherein the keyword is based on a name of the user, an address of the user, a purchase history of the user, or items recommended for the user.

23. One or more computer-readable media as recited in claim 20, wherein the statistically improbable phrase is determined to be associated with the keyword by analyzing the index.

24. One or more computer-readable media as recited in claim 19, further storing computer-executable instructions that, when executed on the one or more processors, perform acts comprising:

receiving a selection of a phrase from the user; and

responsive to the receiving of the selection, associating the selected phrase with the user.

25. A system comprising:

one or more processors; and

the one or more computer-readable media storing the computer-executable instructions as recited in claim 19.

26. A method implemented at least in part by a computing device, the method comprising:

determining that a statistically-improbable phrase is associated with a user;

determining phrases that are associated with the statistically-improbable phrase; and

suggesting a phrase of the determined phrases to the user.

27. A method as recited in claim 26, wherein the statistically improbable phrase is determined to be associated with the user based on known information about the user or prior behavior of the user.

28. A method as recited in claim 26, wherein the statistically improbable phrase is determined to be associated with the user based on a keyword that is manually inputted by the user.

29. A method as recited in claim 26, wherein the determining phrases that are associated with the statistically improbable phrase comprises:

determining one or more words that compose the statistically improbable phrase;

inputting, into an index that maps keywords to phrases, the determined one or more words of the statistically improbable phrase as keywords to the index; and

determining phrases that are associated with one or more of the inputted keywords.

30. A method as recited in claim 26, wherein the suggesting of the phrase to the user comprises suggesting a phrase that is more highly related to the statistically improbable phrase than another, non-suggested phrase that is also associated with the statistically improbable phrase.

31. A method as recited in claim 26, further comprising associating the suggested phrase with the user and with a payment instrument of the user in response to receiving a selection of the suggested phrase.

32. A system comprising:

one or more processors;

memory;

a Pseudo-SIP module, stored in the memory and executable on the one or more processors to determine phrases that are associated with a statistically improbable phrase; and

a phrase-suggestion module, stored in the memory and executable on the one or more processors to output a portion of the determined phrases for use as identifiers.

33. A system as recited in claim 32, wherein the Pseudo-SIP module is further executable on the one or more processors to:

determine one or more words that compose the statistically improbable phrase;

input, into an index that maps keywords to phrases, the determined one or more words of the statistically improbable phrase as keywords to the index; and

determine phrases that are associated with one or more of the inputted keywords.

34. A system as recited in claim 32, wherein the portion of the determined phrases outputted by phrase-suggestion module comprises phrases that are more highly-related to the statistically improbable phrase than other phrases that are also associated with the statistically improbable phrase.