Generating a score of a consumer correction submission

• US 8,417,562 B1
• Filed: 06/11/2008
• Issued: 04/09/2013
• Est. Priority Date: 06/15/2007
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Siemens, Ronald G.
• Primary Examiner(s): Gart, Matthew
• Assistant Examiner(s): Danzig, Reva R
• Application Number:
  US12/137,481
• Time in Patent Office: 1,763 Days
• Field of Search: 705/1, 705/26.1, 707/5, 707/100, 706/12, 706/50, 706/60
• US Class Current: 705/7.38
• CPC Class Codes: G06Q 10/0635 Risk analysis of enterprise or organisation activities
• CPC Class Codes: G06Q 30/0218 based on score

Abstract

Systems and methods for generating a score for use in automated processing of correction submissions are presented. In one embodiment, a submission value from a correction submission obtained from a submitter is obtained. The submission value corresponds to a proposed correction of an attribute of an item, the item having a plurality of attributes that describe the item, and the submission value corresponds to one of the plurality of attributes of the item. The current attribute value of the item that corresponds to the submission value is obtained. A first set of characterization values is generated for the submission value according to a first set of criteria. A second set of characterization values for the current attribute value is generated according to a second set of criteria. Using the first and second sets of characterization values, a score is generated for the correction submission. The score is stored in a data store in association with the correction submission.

Claims

1. A method for generating a risk score for use in automated processing of correction submissions, the method comprising:

performing by one or more computers;

receiving correction information specifying a proposed correction of a specific attribute of an item description for an item, the item description including a plurality of attributes that describe the item including said specific attribute;

generating a first set of one or more characterization values for the proposed correction to the specific attribute according to a first set of one or more criteria;

generating a second set of one or more characterization values for the current value of the specific attribute according to a second set of one or more criteria;

generating one or more difference values between one or more corresponding characterization values of the first and the second sets of one or more characterization values;

based on the one or more difference values, generating a risk score for the proposed correction, the risk score quantifying a level of risk associated with automatically accepting the proposed correction to the specific attribute; and

storing the risk score in a data store.

2. The method of claim 1, further comprising:

performing one or more summation operations on the first and second sets of one or more characterization values in order to generate the risk score.

3. The method of claim 2, wherein there is a one-to-one correspondence between the characterization values of the first and second sets of characterization values.

4. The method of claim 2,

wherein said one or more summation operations are also performed on the one or more difference values.

5. The method of claim 1,

wherein said generating the first set of the one or more characterization values comprises determining a number of words or phrases in the proposed correction that are found in a trouble words data store, and

wherein said generating the second set of the one or more characterization values comprises determining a number of words or phrases in the current value of the specific attribute that are found in the trouble words data store.

6. The method of claim 1,

wherein said generating the first set of the one or more characterization values comprises determining a number of words or phrases in the proposed correction that are found in an offensive words data store, and

wherein said generating the second set of the one or more characterization values comprises determining a number of words or phrases in the current value of the specific attribute that are found in the offensive words data store.

7. The method of claim 1,

wherein said generating the first set of the one or more characterization values comprises determining a number of words in the proposed correction, and

wherein said generating the second set of the one or more characterization values comprises determining a number of words or phrases in the current value of the specific attribute.

8. A method for generating a risk score for use in automated processing of correction submissions, the method comprising:

performing, by one or more computers;

receiving correction information specifying a proposed correction of a specific attribute of an item description for an item, the item description including a plurality of attributes that describe the item including said specific attribute;

generating a first set of one or more characterization values for the proposed correction to the specific attribute according to a first set of one or more criteria;

generating a second set of one or more characterization values for the current value of the specific attribute according to a second set of one or more criteria;

generating one or more difference values between one or more corresponding characterization values of the first and the second sets of one or more characterization values, wherein the one or more difference values are each associated with a comparison between the current value of said specific attribute and said proposed correction;

based on the one or more difference values, generating a risk score for the proposed correction, the risk score quantifying a level of risk associated with automatically accepting the proposed correction to the specific attribute; and

storing the risk score in a data store.

9. The method of claim 8, further comprising:

performing one or more summation operations on the first and second sets of one or more characterization values and the one or more difference values in order to generate the risk score.

10. The method of claim 9, wherein there is a one-to-one correspondence between the characterization values of the first and second sets of characterization values.

11. A computer-implemented system for generating a risk score for use in automated processing of correction submissions, the system comprising:

a computing device having an executable scoring module which generates a risk score for a correction submission from a submitter;

a description data store storing a plurality of item descriptions; and

a submitter profile data store storing information regarding a plurality of correction submitters;

wherein the scoring module when executed by one or more processors of the computing device is configured to;

receive the correction submission from the submitter, the correction submission specifying a proposed correction of a specific attribute of an item'"'"'s item description stored in the description data store, the item description including a plurality of attributes that describe the item including said specific attribute;

generate a first set of one or more characterization values for the proposed correction to the specific attribute according to a first set of one or more criteria;

generate a second set of one or more characterization values for the current value of the specific attribute according to a second set of one or more criteria;

generate one or more difference values between one or more corresponding characterization values of the first and the second sets of one or more characterization values;

based on the one or more difference values, generate a risk score for the proposed correction, the risk score quantifying a level of risk associated with automatically accepting the proposed correction to the specific attribute; and

store the risk score in the description data store.

12. The system of claim 11, wherein the scoring module is further configured to:

perform one or more summation operations on the first and second sets of one or more characterization values in order to generate the risk score.

13. The system of claim 12,

wherein said one or more summation operations are also performed on the one or more difference values.

14. A computer-implemented system for generating a risk score for use in automated processing of correction submissions, the system comprising:

a computing device having an executable scoring module which generates a risk score for a correction submission from a submitter;

a description data store storing a plurality of item descriptions; and

a submitter profile data store storing information regarding a plurality of correction submitters;

wherein the system is configured to;

receive correction information specifying a proposed correction of a specific attribute of an item description for an item, the item description including a plurality of attributes that describe the item including said specific attribute;

generate a first set of one or more characterization values for the proposed correction to the specific attribute according to a first set of one or more criteria;

generate a second set of one or more characterization values for the current value of the specific attribute according to a second set of one or more criteria;

generate one or more difference values between one or more corresponding characterization values of the first and second sets of one or more characterization values, wherein the one or more difference values are each associated with a comparison between the current value of said specific attribute and said proposed correction;

based on the one or more difference values, generate a risk score for the proposed correction, the risk score quantifying a level of risk associated with automatically accepting the proposed correction to the specific attribute; and

store the risk score in the description data store.

15. A non-transitory computer-readable medium bearing computer-executable instructions which, when executed on a computing device, carry out a method for generating a score for use in automated processing of correction submissions, the method comprising:

receiving correction information specifying a proposed correction of a specific attribute of an item description for an item, the item description including a plurality of attributes that describe the item including said specific attribute;

generating a first set of one or more characterization values for the proposed correction to the specific attribute according to a first set of one or more criteria;

generating a second set of one or more characterization values for the current value of the specific attribute according to a second set of one or more criteria;

generating one or more difference values between one or more corresponding characterization values of the first and the second sets of one or more characterization values;

based on the one or more difference values, generating a risk score for the proposed correction, the risk score quantifying a level of risk associated with automatically accepting the proposed correction to the specific attribute; and

storing the risk score in a data store.