System and method for automated hybrid sequencing database generation

Patent metadata

Field	Value
Publication number	US-10783185-B2
Application number	US-201715691563-A
Country	US
Kind code	B2
Filing date	Aug 30, 2017
Priority date	Aug 30, 2017
Publication date	Sep 22, 2020
Grant date	Sep 22, 2020

Abstract

Systems and methods for automated sequencing database generation are disclosed herein. The system can include memory that can include a content library database; a graph database; and a model database. The system can include a user device and at least one server. The at least one server can: receive a content aggregation from the content library database; identify content components of the content aggregation based on a natural language processing analysis of at least a portion of the content aggregation; identify explicit sequencing of the content components; generate an intermediate content graph based on the explicit sequencing of the content components; generate a final content graph from the intermediate content graph based on implicit sequencing of the content components; and store the final content graph within the graph database.

First claim

What is claimed is:

1. A system for generating a hybrid knowledge graph database, the system comprising:
memory comprising:
a content library database comprising at least one content aggregation for presentation to a user;
a graph database comprising:
a first content graph comprising a first intermediate content graph organized according to explicit sequencing and a first final content graph organized according to explicit sequencing and implicit sequencing; and
a second content graph comprising a second intermediate content graph and a second final content graph; and

a model database comprising at least one statistical model;

a user device; and
at least one server, wherein the at least one server is configured to:
receive the first content graph and the second content graph, wherein each of the first content graph and the second content graph comprise a pedagogical graph associated with a content aggregation representing a body of knowledge;
automatically align the first content graph and the second content graph according to a machine learning algorithm, wherein aligning comprises:
identifying first nodes within the first content graph and second nodes within the second content graph;
identifying matching nodes between the first and second content graphs; and
identifying nodes contained in one of the first and second content graphs and missing in the other of the first and second content graphs; and

generate a hybrid content graph based on the aligned first and second content graphs, the hybrid content graph comprising a plurality of nodes, wherein generating the hybrid content graph comprises determining prerequisite relationships between nodes in each of the first and second content graphs and generating sequential relationships between the nodes in the hybrid content graph based on features extracted from each of the first and second content graphs, and wherein the sequential relationships between the nodes in the hybrid content graph are generated according to a combination of an explicit sequencing and an implicit sequencing of content associated with each of the first content graph and the second content graph, wherein the implicit sequencing is determined based on metadata of the content aggregation, wherein the metadata comprises at least one of: back matter and front matter; and
store the hybrid content graph in the graph database.

2. The system of claim 1 , wherein the plurality of nodes of the hybrid content graph are associated with portions of the content aggregation, wherein the hybrid content graph comprises a plurality of edges, and wherein each of the edges links a pair of nodes from the plurality of nodes in a sequential relationship.

3. The system of claim 2 , wherein the plurality of nodes in the hybrid content graph comprise the matching nodes and at least some of the nodes contained in one of the first and second content graphs and missing in the other of the first and second content graphs.

4. The system of claim 3 , wherein at least some of the sequential relationships between the nodes in the hybrid content graph are different than the sequential relationship between corresponding nodes in one or both of the first and second content graphs.

5. The system of claim 4 , wherein aligning the first content graph and the second content graph comprises analyzing nodes in the first and second content graphs and content associated with those nodes.

6. The system of claim 5 , wherein the nodes and content associated with the nodes of the first and second content graphs are analyzed according to a natural language processing algorithm.

7. The system of claim 6 , wherein content associated with the nodes of the first and second content graphs comprises front matter.

8. The system of claim 1 , wherein aligning the first content graph and the second content graph according to a machine learning algorithm comprises: extracting features from content associated with each of the nodes of the first and second content graphs; inputting the extracted features into a machine learning model; and receiving an output from the machine learning model identifying matching nodes between the first and the second content graphs.

9. The system of claim 8 , wherein aligning the first content graph and the second content graph according to a machine learning algorithm further comprises: identifying sequential relationships between the identified matching nodes.

10. The system of claim 9 , wherein the sequential relationships between the identified matching nodes are identified via a Relational Machine Learning analysis.

11. A method for generating a hybrid knowledge graph database, the method comprising:
receiving with at least one server a first content graph and a second content graph, wherein each of the first content graph and the second content graph comprise a pedagogical graph associated with a content aggregation representing a body of knowledge;
automatically aligning with the at least one server the first content graph and the second content graph according to a machine learning algorithm, wherein aligning comprises:
identifying first nodes within the first content graph and second nodes within the second content graph;
identifying matching nodes between the first and second content graphs; and
identifying nodes contained in one of the first and second content graphs and missing in the other of the first and second content graphs; and
generating with the at least one server a hybrid content graph based on the aligned first and second content graphs, the hybrid content graph comprising a plurality of nodes, wherein generating the hybrid content graph comprises determining prerequisite relationships between nodes in each of the first and second content graphs and generating sequential relationships between the nodes in the hybrid content graph based on features extracted from each of the first and second content graphs, and wherein the sequential relationships between the nodes in the hybrid content graph are generated according to a combination of an explicit sequencing and an implicit sequencing of content associated with each of the first content graph and the second content graph, wherein the implicit sequencing is determined based on metadata of the content aggregation, wherein the metadata comprises at least one of: back matter and front matter; and
storing with the at least one server the hybrid content graph in the graph database.

12. The method of claim 11 , wherein the plurality of nodes of the hybrid content graph are associated with portions of the content aggregation, wherein the hybrid content graph comprises a plurality of edges, and wherein each of the edges links a pair of nodes from the plurality of nodes in the sequential relationship.

13. The method of claim 12 , wherein the plurality of nodes in the hybrid content graph comprise the matching nodes and at least some of the nodes contained in one of the first and second content graphs and missing in the other of the first and second content graphs.

14. The method of claim 13 , wherein at least some of the sequential relationships between the nodes in the hybrid content graph are different than the prerequisite relationships between corresponding nodes in one or both of the first and second content graphs.

15. The method of claim 14 , wherein aligning the first content graph and the second content graph comprises analyzing nodes in the first and second content graphs and content associated with those nodes according to a natural language processing algorithm.

16. The method of claim 11 , wherein aligning the

Assignees

• Pearson Education Inc

Inventors

• Murray William
• Baikadi Alok

Classifications

• G06N5/01

  Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound · CPC title

• G06N7/01

  Probabilistic graphical models, e.g. probabilistic networks · CPC title

• G06N5/022 Primary

  Knowledge engineering; Knowledge acquisition · CPC title

• G06N20/00

  Machine learning · CPC title

• G06F16/9024 Primary

  Graphs; Linked lists (G06F16/9027 takes precedence) · CPC title