Detection of unauthorized information exchange between users

• US 9,870,713 B1
• Filed: 09/17/2012
• Issued: 01/16/2018
• Est. Priority Date: 09/17/2012
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Ducrou, Jon Robert, Smith, Brandon J., Sanders, III, Kenneth O., Barry, Marcus A., Brennan, Christopher M., Lerner, David M., Miller, Erik J.
• Primary Examiner(s): Yao, Sam
• Assistant Examiner(s): Humphrey, Michael
• Application Number:
  US13/621,664
• Time in Patent Office: 1,947 Days
• Field of Search: 434362
• CPC Class Codes: G06Q 50/20 Education
• CPC Class Codes: G09B 5/08 providing for individual presentation of information to a plurality of student stations
• CPC Class Codes: G09B 7/07 providing for individual presentation of questions to a plurality of student stations
• CPC Class Codes: G09B 7/12 characterised by modifying the teaching programme in response to a wrong answer, e.g. repeating the question, supplying further information

Abstract

Detection and prevention of unauthorized exchange of information between users of user devices is disclosed. Monitoring data associated with respective information received by the user devices from respective users may be received. Position information indicating proximity or relative positions of the devices may also be received. The monitoring data and/or the position information may be analyzed to determining whether unauthorized information exchange between users may be occurring. One or more metrics indicating a likelihood that unauthorized information exchange is occurring may be generated based on the analysis of the monitoring data and/or the position information. A generated metric may be compared to an associated threshold and preventive measures for deterring unauthorized information exchange may be identified based on a result of the comparison.

Claims

1. A method, comprising:

initiating, by one or more computer processors coupled to at least one memory, a first camera at a first user device;

determining, by one or more computer processors and using the first camera, first behavioral data comprising a first duration of time that a first user'"'"'s eyes deviated from the first user device;

receiving first monitoring data associated with a first set of responses received by the first device from the first user in response to a set of questions, wherein the first monitoring data comprises at least;

the first behavioral data; and

a first viewing time period corresponding to a first length of time that a first question was viewed by the first user;

receiving second monitoring data associated with a second set of responses received by a second device from a second user in response to the set of questions, wherein the second monitoring data comprises at least;

second behavioral data comprising a second duration of time that the second user'"'"'s eyes deviated from the second user device; and

a second viewing time period corresponding to a second length of time that the first question was viewed by the second user;

determining, by one or more computer processors, based on an analysis of the first monitoring data and the second monitoring data, a similarity metric indicative of a degree of correspondence between the first monitoring data and the second monitoring data; and

generating, by one or more computer processors, a cheating metric based on the similarity metric, wherein the cheating metric indicates a probability that a portion of the first set or the second set of responses was obtained from an unauthorized source.

2. The method of claim 1, wherein each of the first monitoring data and the second monitoring data comprises respective data indicating a number or identity of questions for which corresponding correct responses or incorrect responses are received from the respective user.

3. The method of claim 2, wherein at least one of:

the first monitoring data comprises the identity of questions for which incorrect responses are included in the first set of responses and the second monitoring data comprises the identity of questions for which incorrect responses are included in the second set of responses, or

the first monitoring data comprises the identity of questions for which correct responses are included in the first set of responses and the second monitoring data comprises the identity of questions for which correct responses are included in the second set of responses.

4. The method of claim 1, wherein at least one of the first monitoring data or the second monitoring data comprises respective user device interaction data associated with one or more detected actions of the first user or the second user.

5. The method of claim 4, further comprising analyzing, by the one or more computers, at least one of:

the respective user device interaction data of the first monitoring data or the respective user device interaction data of the second monitoring data to generate an analysis result, wherein the cheating metric is further determined based at least in part on the analysis result.

6. The method of claim 4, wherein the one or more detected actions comprise at least one of:

deviation of a line of sight associated with the first user or a line of sight associated with the second user from a respective reference point by more than a respective deviation threshold, or

rotation of the first device or rotation of the second device by more than a respective threshold angle from a respective reference axis.

7. The method of claim 1, further comprising:

obtaining, by the one or more computers, position information indicating a position of the first device relative to the second device.

8. The method of claim 7, further comprising:

generating, by the one or more computers, the cheating metric further based at least in part on the position information.

9. The method of claim 1, wherein the unauthorized source comprises at least one of:

the second user, the second set of responses, or other information obtained by the first user in an unauthorized manner.

10. The method of claim 1, further comprising:

determining, by the one or more computers, that the cheating metric exceeds a threshold value; and

selecting, by the one or more computers, a first set of additional questions for presentation to the first user and a second set of additional questions for presentation to the second user in accordance with one or more preventive measures.

11. The method of claim 10, wherein the one or more preventive measures comprise at least one of:

randomizing an order in which the first set of additional questions are to be presented to the first user or an order that the second set of additional questions are to be presented to the second user,

randomizing, for at least one question common to the first set of additional questions and the second set of additional questions, an order in which a respective set of answer choices are to be presented to the first user or the second user, or

selecting a first question for inclusion in the first set of additional questions and a second question for inclusion in the second set of additional questions from different pools of questions, wherein respective difficulty levels associated with the first question and the second question are substantially the same.

12. The method of claim 10, further comprising:

providing, by the one or more computers, a third user device with access to at least one of;

the cheating metric or the one or more preventive measures.

13. The method of claim 1, further comprising:

receiving, by the one or more computers, at least one of;

first historical usage data associated with the first user or second historical usage data associated with the second user.

14. The method of claim 13, wherein the cheating metric is generated further based at least in part on at least one of:

the first historical usage data or the second historical usage data.

15. A system comprising:

a first user device with a first camera;

a second user device with a second camera; and

one or more computers in communication with the first user device and the second user device, the one or more computers comprising;

at least one processor; and

at least one memory storing computer-executable instructions, wherein at least one processor of the one or more processors is configured to execute the computer-executable instructions to;

determine first behavioral information using the first camera, the first behavioral information comprising a first duration of time that a first user'"'"'s eyes deviated from the first user device;

determine second behavioral information using the second camera, the second behavioral information comprising a second duration of time that a second user'"'"'s eyes deviated from the second user device;

receive first monitoring data associated with first response information received by the first device from the first user, wherein the first monitoring data comprises the first behavioral information and a first viewing time period corresponding to a first length of time that a first question was viewed by the first user;

receive second monitoring data associated with second response information received by the second device from the second user, wherein the second monitoring data comprises the second behavioral information and a second viewing time period corresponding to a second length of time that the first question was viewed by the second user;

analyze the first monitoring data and the second monitoring data to identify one or more correlations; and

generate, based at least in part on the one or more correlations, a cheating metric indicating a probability that a portion of the first or the second response information was obtained from an unauthorized source.

16. The system of claim 15, wherein the at least one processor is further configured to execute the computer-executable instructions to:

determine that the cheating metric exceeds a threshold value; and

generate one or more preventive measures based on the determination that the cheating metric exceeds the threshold value, wherein the one or more preventive measures modify one or more characteristics associated with question information presented to the first user or one or more characteristics associated with question information presented to the second user.

17. The system of claim 15, wherein the cheating metric is a first cheating metric indicating a probability that the first response information was obtained from the first unauthorized source, and wherein the at least one processor is further configured to execute the computer-executable instructions to:

generate, based at least in part the one or more correlations, a second cheating metric indicating a probability that the second response information was obtained from the second unauthorized source; and

generate one or more preventive measures based at least in part on a deviation between the first cheating metric and the second cheating metric,

wherein the one or more preventive measures modify one or more characteristics associated with question information presented to the first user or one or more characteristics associated with question information presented to the second user.

18. The system of claim 17, wherein the at least one processor is further configured to execute the computer-executable instructions to:

implement the one or more preventative measures in real-time during administration of an examination with which the first response information and the second response information are associated.

19. The system of claim 15, wherein the at least one processor is further configured to execute the computer-executable instructions to determine the first length of time of the first viewing time period by:

receiving a first timestamp associated with initial presentation of the first question;

receiving a second timestamp associated with submission of a response to the first question; and

determining a difference between the first timestamp and the second timestamp.

20. The system of claim 19, wherein the at least one processor is further configured to execute the computer-executable instructions to determine the first length of time of the first viewing time period by:

receiving a third timestamp associated with presentation of a second question;

determining that the first question is unanswered; and

receiving a fourth timestamp associated with a second presentation of the first question.