Educational and content recommendation management system
An educational system may include a contextualizer that accesses and processes content items and build a knowledge base including educational content and relationships between content items. The system may include a user assessor unit that determines user assessment variables based on user responses to previous educational content items and generate a user model based on the user assessment variables. The system may include a recommender that navigates the knowledge base and generates recommendations of new educational content based on the user model. The recommendation can assess the user's proficiency on a given concept based on the user model and maximize the user's probability of success around the concept of which the user shows proficiency, or improve the user's proficiency around the concept the user shows weakness. In some examples, the system may acquire an expert's knowledge about certain concepts to refine the user model, increasing the accuracy of the recommender.
1. A system for delivering educational content to a user, the system comprising:
a contextualizer configured to access a content database comprising a plurality of content items, wherein the contextualizer:
generates a knowledge base based on the content items, wherein the knowledge base comprises a plurality of nodes and a plurality of edges extending between the plurality of odes, the plurality of nodes and the plurality of edges corresponding to relationships between the plurality of content items; and
generates an update to the knowledge base in response to a new content item being added to the plurality of content items, wherein the update to the knowledge base includes inferring a position of the new content item in the knowledge base based on the plurality of nodes and plurality of edges generated based on the content database without the new content item;
a user knowledge assessor configured to receive one or more user assessment variables based on a user's response to a first content item of a content database, the content database comprising a plurality of content items; and
a recommender in communication with the user knowledge assessor and configured recommend a second content item for displaying to the user, based on the one or more user assessment variables.
2. The system of claim 1 , wherein the contextualizer is further configured to generate the update to the knowledge space without recontexualizing the knowledge base.
3. The system of claim 2 , wherein:
generating the knowledge base further comprises:
access the plurality of content items from the content database;
form a corpus based on plurality of the content items;
generate a probability model of the content items; and
use the probability model to generate the knowledge base by:
generating the plurality of nodes of knowledge base, each node representing a topic; and
generating the plurality of edges amongst the nodes of the graph, each edge of the plurality of edges connecting a first node of the plurality of nodes and a second node of the plurality of nodes and including a weight representing a probability of the first node of the plurality of nodes and the second node of the plurality of nodes belonging to a same concept.
4. The system of claim 3 , wherein the probability model is a latent Dirichlet allocation (LDA) model.
5. The system of claim 2 , wherein the recommender is further configured to:
access the knowledge base;
generate a user model based on the one or more user assessment variables;
use the user model to navigate the knowledge base to determine a node in the knowledge base;
access content represented by the determined node; and
recommend the second content item based on the content.
6. The system of claim 5 , wherein the one or more user assessment variables comprise one or more of a veracity related to the first content item, a response time of the user responding to the first content item, or a confidence value of the user responding to the first content item.
7. The system of claim 5 , wherein the recommender is further configured to:
determine a user score based on the user's response to at least the first content item and the second content item; and
recommend a third content item associated with a concept different from a concept associated with the second content item, if the user score has reached a threshold.
8. The system of claim 5 , wherein the recommender is further configured to determine the second content item from a plurality of candidate content items, wherein the second content item is associated with a maximum probability of success among the candidate content items, and wherein the probability of success associated with a respective content item represents a probability that the user will respond to the respective content item correctly.
9. The system of claim 3 , wherein generating the update to the knowledge space further comprises determining the position of the new content item based on the first node of the plurality of nodes and the second node of the plurality of nodes, such that the new content item does not change the first node of the plurality of nodes or the second node of the plurality of nodes.
10. The system of claim 5 , wherein the contextualizer is further configured to preserve the user model based on the knowledge base after the new content item is added to the knowledge base.
11. A contextualizer in an educational system comprising:
a processing device; and
computer readable medium containing programming instructions that, when executed, will cause the processing device to:
access original content items;
form a corpus based on the original content items;
generate probability model of original content items;
use the probability model to generate a first knowledge base by:
generating a plurality of nodes each representing a topic of the original content;
generating edges amongst the plurality of nodes, each edge connecting a first node and a second node and including a weight representing a probability of the first node and the second node belonging to a same topics access a new content item;
determine a position of the new content item in the knowledge base using the plurality of nodes generated based on the original content items; and
analyze a user response to engagement with a first content item; and
presenting a second content item based on the user response to the first content item and a position of the first content item in the knowledge base.
12. The system of claim 11 , wherein the position of the new content item does not change the first node or the second node in the knowledge base.
13. The method of claim 11 , wherein the user response is analyzed based on receiving one or more user assessment variables after or during presentation of the first content item to the user.
14. A method for delivering educational content to a user comprising:
forming a corpus based on first content items;
generating a probability model of relationships of the first content items;
using the probability model to contextualize a knowledge base by:
generating nodes, wherein each node represents a topic within the first content items; and
generating edges between the nodes, each edge connects a first node and a second node and includes a weight representing a probability that the first node and the second node belong to a same topic;
access a second content item; and
adding the second content item to the knowledgebase without recontextualizing the knowledge base; and
presenting the first content item or the second content item for display to a user based on a location of the first content item and the second content item in the knowledge base.
15. The method of claim 14 , wherein when generating edges amongst the nodes of the graph, each edge connecting the first and second node including the weight representing the probability of the first node and the second node belonging to a related concept.