Mining phrases for association with a user

• US 8,706,644 B1
• Filed: 01/13/2009
• Issued: 04/22/2014
• 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: Jesensky, James, Oates, Isaac, Huynh, Steve, Vaidya, Vinay
• Primary Examiner(s): Smits, Talivaldis Ivars
• Assistant Examiner(s): Roberts, Shaun
• Application Number:
  US12/353,059
• Time in Patent Office: 1,925 Days
• Field of Search: 705/74, 705/44, 705/67, 704/1, 704/9, 726/6
• US Class Current: 705/74
• CPC Class Codes: G06Q 20/382 insuring higher security of transaction
• CPC Class Codes: G06Q 20/401 Transaction verification

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 for executing a transaction. 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. One or more non-transitory computer-readable media storing computer-executable instructions that, when executed on one or more processors, perform acts comprising:

analyzing individual ones of multiple corpuses of text to determine statistically improbable phrases that individually appear in at least one of the corpuses of text more than a threshold number of times;

storing the determined statistically improbable phrases in a first corpus of phrases, wherein the statistically improbable phrases stored in the first corpus of phrases include a verb phrase;

parsing individual ones of the multiple corpuses of text to determine phrases comprising two or more contiguous words within the text;

storing the determined phrases comprising two or more contiguous words in the first corpus of phrases;

filtering the first corpus of phrases to define a second corpus of phrases comprising fewer phrases than the first corpus of phrases;

causing phrases of the second corpus of phrases to be suggested to a user;

receiving a selection of a phrase of the suggested phrases from the user; and

associating the selected phrase with the user and with one or more of a payment instrument associated with the user, a location associated with the user, or a delivery method associated with the user.

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

responsive to determining that a corpus of text includes a statistically improbable phrase, storing the determined statistically improbable phrase in a corpus of phrases for association with a user account, wherein the statistically improbable phrase, when received before or during a transaction, provides access to a payment instrument for use in a transaction;

and causing output of the stored statistically improbable phrase for association with a user account.

3. A method as recited in claim 2, further comprising:

suggesting the statistically improbable phrase to a user of the user account; and

responsive to receiving a selection of the statistically improbable phrase from the user,

associating the statistically improbable phrase with the user account.

4. A method as recited in claim 3, wherein the statistically improbable phrase is associated with the payment instrument of the user in response to the receiving of the selection of the statistically improbable phrase from the user.

5. A method as recited in claim 2, wherein the statistically improbable phrase comprises between two and seven grammatically-correct words.

6. A method as recited in claim 2, further comprising:

parsing the corpus of text to determine phrases comprising two or more contiguous words within the text; and

storing the determined phrases comprising two or more contiguous words in the corpus of phrases for suggestion to a user of the user account.

7. A method as recited in claim 2, further comprising:

determining at least one keyword associated with a user of the user account;

determining phrases of the corpus of phrases that are associated with the at least one keyword;

if the phrases determined to be associated with the at least one keyword include the statistically improbable phrase, suggesting the statistically improbable phrase to the user; and

otherwise, suggesting one or more phrases other than the statistically improbable phrase to the user.

8. A method as recited in claim 2, wherein the corpus of text comprises a book, a magazine, online content, audio content, or video content.

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

determining that a source includes a statistically improbable phrase; and

based at least in part on determining that the source includes a statistically improbable phrase, storing the determined statistically improbable phrase in a corpus of phrases for use as an identifier, wherein the statistically improbable phrase, when received before or during a transaction, provides access to a payment instrument for use in a transaction.

10. A method as recited in claim 9, further comprising outputting the statistically improbable phrase for selection by a user.

11. A method as recited in claim 9, further comprising:

outputting the statistically improbable phrase for selection by a user;

receiving a selection of the statistically improbable phrase by the user; and

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

12. A method as recited in claim 9, further comprising:

outputting the statistically improbable phrase for selection by a user;

receiving a selection of the statistically improbable phrase by the user; and

responsive to the receiving of the selection, associating the statistically improbable phrase with the user and with a payment instrument associated with the user, a location associated with the user, or a delivery method associated with the user.

13. A method as recited in claim 9, wherein the corpus of phrases includes multiple other phrases for use as identifiers, at least some of the multiple other phrases comprising two or more contiguous words found within the source.

14. A method as recited in claim 9, further comprising:

determining that multiple other sources individually include one or more statistically improbable phrases; and

storing individual ones of the determined statistically improbable phrases in the corpus of phrases, individual ones of the stored statistically improbable phrases for use as an identifier.

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

parsing a corpus of text to determine phrases comprising two or more contiguous words within the text;

storing the determined phrases comprising two or more contiguous words in a first corpus of phrases, wherein at least one word of the two or more contiguous words is a preposition;

filtering out a portion of the first corpus of phrases to define a second corpus of phrases; and

providing phrases of the second corpus of phrases for use as identifiers, wherein the phrases, when received before or during a transaction, provides access to a payment instrument for use in a transaction.

16. One or more non-transitory computer-readable media as recited in claim 15, wherein the corpus of text comprises a book, and wherein the two or more contiguous words comprise words that are directly contiguous in the book and free from separation by punctuation.

17. One or more non-transitory computer-readable media as recited in claim 15, wherein the filtering comprises:

determining, for individual words of individual phrases, a part-of-speech associated with the word;

determining, based on the determined part-of-speech for the individual words, a part-of-speech combination of the individual phrases; and

comparing the determined part-of-speech combination of the individual phrases with a part-of-speech list.

18. One or more non-transitory computer-readable media as recited in claim 17, wherein the part-of-speech list comprises a list that specifies part-of-speech combinations for exclusion, and wherein the filtering further comprises filtering out phrases having determined part-of-speech combinations that appear on the list.

19. One or more non-transitory computer-readable media as recited in claim 17, wherein the part-of-speech list comprises a list that specifies part-of-speech combinations for inclusion, and wherein the filtering further comprises filtering out phrases having determined part-of-speech combinations that do not appear on the list.

20. A system comprising:

one or more processors; and

the one or more non-transitory computer-readable media storing the computer-executable instructions configured to;

parse a corpus of text to determine phrases comprising two or more contiguous words within the text;

store the determined phrases comprising two or more contiguous words in a first corpus of phrases, wherein at least one word of the two or more contiguous words is a preposition;

filter out a portion of the first corpus of phrases to define a second corpus of phrases; and

provide phrases of the second corpus of phrases for use as identifiers, wherein the phrases, when received before or during a transaction, provides access to a payment instrument for use in a transaction.

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

parsing a source to determine phrases comprising two or more contiguous words within the source;

storing the determined phrases, the words of the stored phrases being different parts of speech; and

associating at least one of the determined phrases comprising two or more contiguous words with a payment instrument, the at least one of the determined phrases, when received before or during a transaction, providing access to the payment instrument for use in a transaction.

22. One or more non-transitory computer-readable media as recited in claim 21, wherein the parsing of the source further comprises disregarding at least some phrases in the source that do not comprise two or more contiguous words.

23. One or more non-transitory computer-readable media as recited in claim 21, wherein the parsing further comprises:

parsing the source to determine if the source includes a statistically improbable phrase; and

if the source includes a statistically improbable phrase, storing the statistically improbable phrase for association with a user account.

24. One or more non-transitory computer-readable media as recited in claim 23, further comprising:

suggesting the statistically improbable phrase to a user;

receiving, from the user, a selection of the statistically improbable phrase; and

responsive to receiving the selection from the user, associating the statistically improbable phrase with a user account associated with the user.

25. One or more non-transitory computer-readable media as recited in claim 21, further storing computer-executable instructions that, when executed on the one or more processors, perform acts comprising:

determining a keyword associated with a user having a user account;

determining phrases associated with the determined keyword, the phrases including at least one of the phrases comprising two or more contiguous words within the source;

suggesting, to the user, the phrases that are determined to be associated with the keyword;

receiving, from the user, a selection of a suggested phrase; and

associating the selected phrase with the user account.

26. A system comprising:

one or more processors; and

the one or more non-transitory computer-readable media storing the computer-executable instructions configured to;

parse a source to determine phrases comprising two or more contiguous words within the source;

store the determined phrases, the words of the stored phrases being different parts of speech; and

associate at least one of the determined phrases comprising two or more contiguous words with a payment instrument, the at least one of the determined phrases, when received before or during a transaction, providing access to the payment instrument for use in a transaction.

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

mining multiple corpuses of text to locate phrases, individual ones of the phrases comprising at least two grammatically-correct words;

storing the located phrases in a corpus of phrases, wherein the located phrases have at least a first word and a second word, the first word being a different part-of-speech than the second word; and

associating at least some of the located phrases with users and corresponding payment instruments associated with the users, locations associated with the users, or delivery methods associated with the users.

28. One or more non-transitory computer-readable media as recited in claim 27, further storing computer-executable instructions that, when executed on the one or more processors, perform acts comprising:

mining the multiple corpuses of text to locate statistically improbable phrases that appear in at least one of the corpuses of text more than a threshold number of times;

storing the located statistically improbable phrases in the corpus of phrases; and

associating at least some of the located statistically improbable phrases with users and corresponding payment instruments associated with the users, locations associated with the users, or delivery methods associated with the users.

29. A system comprising:

one or more processors; and

the one or more non-transitory computer-readable media storing the computer-executable instructions configured to;

mine multiple corpuses of text to locate phrases, individual ones of the phrases comprising at least two grammatically-correct words;

store the located phrases in a corpus of phrases, wherein the located phrases have at least a first word and a second word, the first word being a different part-of-speech than the second word; and

associate at least some of the located phrases with users and corresponding payment instruments associated with the users, locations associated with the users, or delivery methods associated with the users.

30. A system comprising:

one or more processors;

memory;

a phrase-mining module, stored in the memory and executable on the one or more processors to;

determine that one or more sources include statistically improbable phrases, and mine multiple sources to determine phrases comprising at least two contiguous words that reside within the multiple sources, wherein at least one phrase of the determined phrases includes a prepositional phrase; and

a phrase-suggestion module, stored in the memory and executable on the one or more processors to provide at least some of the statistically improbable phrases and at least some of the phrases comprising at least two contiguous words as identifiers that, when received before or during a transaction, provide access to a payment instrument for use in a transaction.

31. A system as recited in claim 30, further comprising a phrase-association module, stored in the memory and executable on the one or more processors to associate the statistically improbable phrases and the phrases comprising at least two contiguous words to users and payment instruments associated with the users, shipping addresses associated with the users, locations to receive digital items or money specified by the users, or delivery methods specified by the users.

32. A system as recited in claim 30, further comprising:

a phrase-filtering module, stored in the memory and executable on the one or more processors to filter out at least some of the statistically improbable phrases and the phrases comprising at least two contiguous words,

wherein the phrase-suggestion module suggests for association with users statistically improbable phrases and phrases comprising at least two contiguous words that the phrase-filtering module has not filtered out.