Systems and methods for automatic cohort misconception remediation

• US 10,296,841 B1
• Filed: 08/25/2016
• Issued: 05/21/2019
• Est. Priority Date: 08/28/2015
• Status: Active Grant

• Current Assignee: Pearson Education Incorporated (Pearson plc)
• Original Assignee: Pearson Education Incorporated (Pearson plc)
• Inventors: Moriarty, Brian, Potter, Mark
• Primary Examiner(s): Shmatov, Alexey
• Assistant Examiner(s): Buss, Benjamin J
• Application Number:
  US15/247,751
• Time in Patent Office: 999 Days
• Field of Search: None
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks
• CPC Class Codes: H04L 67/1097 for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
• CPC Class Codes: H04L 67/52 specially adapted for the location of the user terminal
• CPC Class Codes: H04L 67/535 Tracking the activity of the user network monitoring arrangements H04L43/00; recording of computer activity G06F11/34
• CPC Class Codes: H04L 67/567 Integrating service provisioning from a plurality of service providers

Abstract

Systems and methods for content selection with first and second recommendation engines are disclosed herein. The system can include a memory include a content library database and a model database. The system can include a user device having a first network interface and a first I/O subsystem. The system can include one or more servers that can include a packet selection system and a presentation system. These one or more servers can: receive response data from the user device; provide received response data to a first recommendation engine; alert a second recommendation engine when a selected next node is a placeholder node; retrieve at least one statistical model relevant to selection of next node content; and select next node content based on an output of the at least one statistical model.

Claims

1. A system for automatic data packet remediation, the system comprising:

memory comprising;

a content library database comprising a plurality of nodes arranged in a content network; and

a threshold database comprising a misconception threshold delineating between an acceptable and an unacceptable probability of an existence of a misconception;

a user device comprising;

a first network interface configured to exchange data via a communication network; and

a first I/O subsystem configured to convert electrical signals to user interpretable outputs via a user interface; and

one or more servers configured to selectively use a rules-based recommendation engine and an adaptive recommendation engine of a dual recommendation engine configuration to perform state-based traversal of at least some of the plurality of nodes and to perform adaptive generation and presentation of node content based at least in part on processing response data received from a plurality of user devices communicatively coupled to the one or more servers, one or more guard conditions mapped to the at least some of the plurality of nodes, and the misconception threshold, based at least in part on;

processing user inputs from the plurality of user devices to identify a user cohort based at least in part on a common attribute, wherein the user cohort comprises a plurality of users sharing the common attribute;

based at least in part on processing data corresponding to at least some of the user inputs with the rules-based recommendation engine, selecting with the rules-based recommendation engine one or more next nodes based at least in part on a current location in the content network, a set of potential next nodes, and at least one of the one or more guard conditions, and selecting a data packet for providing to the user devices associated with the user cohort with the rules-based recommendation engine according to rules specifying data packets based at least in part on inputs corresponding to at least one of the plurality of users, wherein the potential next nodes in the content network are pairwise connected via a plurality of edges and are mapped to the at least one of the one or more guard conditions, and the data packet is mapped to the selected one or more next nodes;

transmitting instances of the selected data packet to the user devices mapped to the user cohort;

receiving responses to the transmitted instances of the data packet from the user devices associated with the user cohort;

translate each of the received responses into a respective observable with a response processor, wherein each observable comprises a respective characterization of the received response;

generating response cohorts comprising groups of users in the user cohort, wherein the response cohorts are generated based on commonalities in the observables of the received responses;

identifying a misconception based on a comparison of a cohort percentage and a misconception threshold;

automatically updating a model trained with the inputs of the at least one the plurality of users with a model engine according to the identified misconception and the observables; and

causing the adaptive recommendation engine to generate an output corresponding to next node content derived from the updated model based at least in part on one or more parameters generated from one or more features of user data associated with the user cohort, and transmitting instances of a different data packet to at least one of the user devices associated with the user cohort to cause presentation of the next node content, the next node content corresponding to remediation content.

2. The system of claim 1, wherein the memory further comprises a user profile database containing data identifying a plurality of users.

3. The system of claim 1, wherein the observable identifies whether the received response comprises a desired response or an undesired response.

4. The system of claim 3, wherein the one or more servers are further configured to generate a plurality of response cohorts, wherein each of the response cohorts comprises a group of users.

5. The system of claim 4, wherein response cohorts are generated based on common attributes of observables generated from received responses.

6. The system of claim 5, wherein the one or more servers are further configured to generate a cohort percentage for each of the response cohorts, wherein the cohort percentage identifies a percentage of users grouped in the response cohort associated with the cohort percentage.

7. The system of claim 6, wherein the one or more servers are further configured to:

retrieve a misconception threshold from the threshold database;

compare the misconception threshold to the cohort percentage for each of the response cohorts; and

identify a potential misconception for one of the response cohorts based on the comparison of the misconception threshold and the cohort percentage of that one of the response cohorts.

8. The system of claim 7, wherein the one or more servers are further configured to update metadata to indicate the identified potential misconception.

9. The system of claim 8, wherein the updated metadata comprises user metadata of users belonging to that one of the response cohorts.

10. The system of claim 8, wherein the updated metadata comprises metadata of the provided data packet.

11. A method for automatic data packet remediation comprising:

selectively using a rules-based recommendation engine and an adaptive recommendation engine of a dual recommendation engine configuration to perform state-based traversal of at least some of a plurality of nodes and to perform adaptive generation and presentation of node content based at least in part on processing response data received from a plurality of user devices communicatively coupled to one or more servers, one or more guard conditions mapped to the at least some of the plurality of nodes, and a misconception threshold, at least in part by;

processing user inputs from the plurality of user devices to identifying a user cohort based at least in part on a common attribute, wherein the user cohort comprises a plurality of user sharing the common attribute;

based at least in part on processing data corresponding to at least some of the user inputs with the rules-based recommendation engine, selecting with the rules-based recommendation engine one or more next nodes based at least in part on a current location in a content network, a set of potential next nodes, and at least one of the one or more guard conditions, and selecting a data packet for providing to the user devices associated with the user cohort with the rules-based recommendation engine according to rules specifying data packets based at least in part on inputs corresponding to at least one of the plurality of users, wherein the potential next nodes in the content network are pairwise connected via a plurality of edges and are mapped to the at least one of the one or more guard conditions, and the data packet is mapped to the selected one or more next nodes;

transmitting instances of the selected data packet to the user devices mapped to the user cohort;

receiving responses to the transmitted instances of the data packet from the user devices associated with the user cohort;

translating each of the received responses with a response processor according to a respective observable, wherein each observable comprises a respective characterization of the received response;

generating response cohorts comprising groups of users in the user cohort, wherein the response cohorts are generated based on commonalities in the observables of the received responses;

identifying a misconception based on a comparison of a cohort percentage and the misconception threshold;

automatically updating a model trained with the inputs of the at least one the plurality of users with a model engine according to the identified misconception and the observables; and

causing the adaptive recommendation engine to generate an output corresponding to next node content derived from the updated model based at least in part on one or more parameters generated from one or more features of user data associated with the user cohort, and transmitting instances of a different data packet to at least one of the user devices associated with the user cohort to cause presentation of the next node content, the next node content corresponding to remediation content.

12. The method of claim 11, wherein generating and sending an alert comprises identifying one or several data packets associated with the identified misconception.

13. The method of claim 11, wherein the observable identifies whether the received response comprises a desired response or an undesired response.

14. The method of claim 13, further comprising generating a plurality of response cohorts, wherein each of the response cohorts comprises a group of users.

15. The method of claim 14, wherein response cohorts are generated based on common attributes of observables generated from received responses.

16. The method of claim 15, further comprising generating a cohort percentage for each of the response cohorts, wherein the cohort percentage identifies a percentage of users grouped in the response cohort associated with the cohort percentage.

17. The method of claim 16, further comprising:

retrieving the misconception threshold from a misconception database;

comparing the misconception threshold to the cohort percentage for each of the response cohorts; and

identifying a potential misconception for one of the response cohorts based on the comparison of the misconception threshold and the cohort percentage of that one of the response cohorts.

18. The method of claim 17, further comprising updating metadata to indicate the identified potential misconception.

19. The method of claim 18, wherein the updated metadata comprises user metadata of users belonging to that one of the response cohorts.

20. The method of claim 18, wherein the updated metadata comprises metadata of the provided data packet.