Internal dataset-based outlier detection for confidential data in a computer system

• US 10,025,939 B2
• Filed: 09/28/2016
• Issued: 07/17/2018
• Est. Priority Date: 09/28/2016
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: Kenthapadi, Krishnaram, Ambler, Stuart MacDonald, Jain, Parul
• Primary Examiner(s): Nipa, Wasika
• Application Number:
  US15/279,334
• Publication Number:
  US 20180089443A1
• Time in Patent Office: 657 Days
• CPC Class Codes: G06F 21/604 Tools and structures for managing or administering access control systems
• CPC Class Codes: G06F 21/6218 to a system of files or objects, e.g. local or distributed file system or database
• CPC Class Codes: G06F 21/6245 Protecting personal data, e.g. for financial or medical purposes

Abstract

In an example, a submission of a confidential data value of a first confidential data type is received from a first user with one or more attributes. A plurality of previously submitted confidential data values of a first confidential data type for a cohort matching the one or more attributes of the first user are retrieved. A plurality of percentiles for the confidential data values are calculated. Then, an interquartile range is calculated for a first and a second of the plurality of percentiles. A lower limit for the first confidential data type and the cohort is computed by taking a maximum of zero or the difference between the value for the first of the plurality of percentiles and a product of a preset alpha parameter and the interquartile range. Then it is determined if the confidential data value submitted by the user is lower than the lower limit.

Claims

1. A system comprising:

one or more hardware processors;

a non-transitory computer-readable medium having instructions stored there on, which, when executed by the one or more hardware processor, cause the system to;

receive, via a first computerized user interface implemented as a screen of a graphical user interface, a submission of a confidential data value of a first confidential data type from a first user, entered into a field of the screen of the graphical user interface;

identify, using the one or more hardware processors, one or more attributes of the first user;

retrieve a plurality of previously submitted confidential data values of a first confidential data type for a cohort matching the one or more attributes of the first user, the previously submitted confidential data values having been encrypted on an external data source, the cohort being a grouping of data pertaining to a combination of user attributes for users who submitted the confidential data values;

calculate, using the one or more hardware processors, a plurality of percentiles for the plurality of previously submitted confidential data values;

calculate, using the one or more hardware processors, an interquartile range for a first and a second of the plurality of percentiles, wherein the value for the first of the plurality of percentiles is lower than the value for the second of the plurality of percentiles;

compute, using the one or more hardware processors, a lower limit for the first confidential data type and the cohort by taking a maximum of zero or a difference between the value for the first of the plurality of percentiles and a product of a preset alpha parameter and the interquartile range;

determine, using the one or more hardware processors, whether the confidential data value submitted by the first user is an outlier by determining if the confidential data value submitted by the user is lower than the lower limit; and

in response to a determination that the confidential data value submitted by the first user is not outlier, permitting, using the one or more hardware processors, the confidential data value submitted by the first user to be used for insights provided to other users.

2. The system of claim 1, wherein the instructions further cause the system to:

compute, using the one or more hardware processors, an upper limit for the first confidential data type by adding the value for the second of the plurality of percentiles to a product of a preset beta parameter and the interquartile range;

wherein the determining whether the confidential data value submitted by the first user is an outlier includes determining if the confidential data value submitted by the user is higher than the higher limit.

3. The system of claim 1, wherein the instructions further cause the system to modify the lower limit by raising the lower limit to a preset minimum value if the preset minimum value is greater than the lower limit.

4. The system of claim 1, wherein the lower limit is computed using Box and Whisker method.

5. The system of claim 1, wherein preventing includes altering the confidential data value to the lower limit if the confidential data value is lower than the lower limit and the first confidential data type is of a first preset type.

6. The system of claim 1, wherein preventing includes altering the confidential data value to the upper limit if the confidential data value is higher than the upper limit and the first confidential data type is of a first preset type.

7. The system of claim 5, wherein the preventing includes removing the confidential data value if the first confidential data type is of a second preset type.

8. A computerized method comprising:

receiving, via a first computerized user interface implemented as a screen of a graphical user interface, a submission of a confidential data value of a first confidential data type from a first user, entered into a field of the screen of the graphical user interface;

identifying, using one or more hardware processors, one or more attributes of the first user;

retrieving, using the one or more hardware processors, a plurality of previously submitted confidential data values of a first confidential data type for a cohort matching the one or more attributes of the first user, the previously submitted confidential data values having been encrypted on an external data source, the cohort being a grouping of data pertaining to a combination of user attributes for users who submitted the confidential data values;

calculating, using the one or more hardware processors, a plurality of percentiles for the plurality of previously submitted confidential data values;

calculating, using the one or more hardware processors, an interquartile range for a first and a second of the plurality of percentiles, wherein the value for the first of the plurality of percentiles is lower than the value for the second of the plurality of percentiles;

computing, using the one or more hardware processors, a lower limit for the first confidential data type and the cohort by taking a maximum of zero or a difference between the value for the first of the plurality of percentiles and a product of a preset alpha parameter and the interquartile range;

determining, using the one or more hardware processors, whether the confidential data value submitted by the first user is an outlier by determining if the confidential data value submitted by the user is lower than the lower limit; and

in response to a determination that the confidential data value submitted by the first user is not outlier, permitting, using the one or more hardware processors, the confidential data value submitted by the first user to be used for insights provided to other users.

9. The method of claim 8, further comprising:

Computing, using the one or more hardware processors, an upper limit for the first confidential data type by adding the value for the second of the plurality of percentiles to a product of a preset beta parameter and the interquartile range;

wherein the determining whether the confidential data value submitted by the first user is an outlier includes determining if the confidential data value submitted by the user is higher than the higher limit.

10. The method of claim 8, further comprising modifying the lower limit by raising the lower limit to a preset minimum value if the preset minimum value is greater than the lower limit.

11. The method of claim 8, wherein the lower limit is computed using Box and Whisker method.

12. The method of claim 8, wherein preventing includes altering the confidential data value to the lower limit if the confidential data value is lower than the lower limit and the first confidential data type is of a first preset type.

13. The method of claim 8, wherein preventing includes altering the confidential data value to the upper limit if the confidential data value is higher than the upper limit and the first confidential data type is of a first preset type.

14. The method of claim 12, wherein the preventing includes removing the confidential data value if the first confidential data type is of a second preset type.

15. A non-transitory machine-readable storage medium comprising instructions, which when implemented by one or more machines, cause the one or more machines to perform operations comprising:

receiving, via a first computerized user interface implemented as a screen of a graphical user interface, a submission of a confidential data value of a first confidential data type from a first user, entered into a field of the screen of the graphical user interface;

identifying, using one or more hardware processors, one or more attributes of the first user;

retrieving, using the one or more hardware processors, a plurality of previously submitted confidential data values of a first confidential data type for a cohort matching the one or more attributes of the first user, the previously submitted confidential data values having been encrypted on an external data source, the cohort being a grouping of data pertaining to a combination of user attributes for users who submitted the confidential data values;

calculating, using the one or more hardware processors, a plurality of percentiles for the plurality of previously submitted confidential data values;

calculating, using the one or more hardware processors, an interquartile range for a first and a second of the plurality of percentiles, wherein the value for the first of the plurality of percentiles is lower than the value for the second of the plurality of percentiles;

computing, using the one or more hardware processors, a lower limit for the first confidential data type and the cohort by taking a maximum of zero or a difference between the value for the first of the plurality of percentiles and a product of a preset alpha parameter and the interquartile range;

determining, using the one or more hardware processors, whether the confidential data value submitted by the first user is an outlier by determining if the confidential data value submitted by the user is lower than the lower limit; and

in response to a determination that the confidential data value submitted by the first user is not outlier, permitting, using the one or more hardware processors, the confidential data value submitted by the first user to be used for insights provided to other users.

16. The non-transitory machine-readable storage medium of claim 15, wherein the operations further comprise:

computing, using the one or more hardware processors, an upper limit for the first confidential data type by adding the value for the second of the plurality of percentiles to a product of a preset beta parameter and the interquartile range;

wherein the determining whether the confidential data value submitted by the first user is an outlier includes determining if the confidential data value submitted by the user is higher than the higher limit.

17. The non-transitory machine-readable storage medium of claim 15, further comprising modifying, using the one or more hardware processors, the lower limit by raising the lower limit to a preset minimum value if the preset minimum value is greater than the lower limit.

18. The non-transitory machine-readable storage medium of claim 15, wherein the lower limit is computed using a box and whisker technique.

19. The non-transitory machine-readable storage medium of claim 15, wherein preventing includes altering the confidential data value to the lower limit if the confidential data value is lower than the lower limit and the first confidential data type is of a first preset type.

20. The non-transitory machine-readable storage medium of claim 15, wherein the preventing includes altering the confidential data value to the upper limit if the confidential data value is higher than the upper limit and the first confidential data type is of a first preset type.