Systems and methods for proctoring and updating courses using machine learning

A system for updating a user interface displaying content related to a topic based on user comprehension of the topic, comprising:

• at least one memory;
• at least one hardware processor coupled with the at least one memory and configured, individually or in combination, to:
• receive, via the user interface (UI), a user selection of the topic;
• generate, using a machine learning algorithm, a course from reference materials describing a plurality of sub-topics associated with the topic, wherein generating the course includes:
• filtering the reference materials to select materials that meet a threshold quality level and a preferred difficulty level,
• executing an unsupervised machine learning model utilizing a graph-based or clustering algorithm to identify and extract selected sentences from the reference materials,
• capping the extracted selected sentences such that a total length of course content is bound to a word limit generated for the sub-topic, and generating a plurality of assessments that test comprehension of the plurality of sub-topics;
• output the course on the UI;
• monitor, in real-time, user interaction with both a first subset of the assessments within the course on the UI and course content tested by the first subset of the assessments;
• determine whether the user interaction meets a comprehension criterion by generating an input vector comprising a plurality of attributes associated with the user interaction and feeding the input vector into a machine learning classifier, wherein the machine learning classifier applies a plurality of learned weights to the input vector to output a classification of whether the user interaction meets the comprehension criterion, the plurality of learned weights being optimized by training the machine learning classifier on a training dataset comprising historical input vectors mapped to corresponding output labels indicating whether historical user interactions satisfy the comprehension criterion;
• in response to determining that the user interaction does not meet the comprehension criterion;
• automatically update the course content displayed in the UI by executing at least one machine learning model configured, individually or in combination, to:
• generate, for an input duration, a new word limit of text in the course content, a media limit of graphics in the course content, and an assessment limit of questions in the course content;
• and
• summarize the reference materials into the course content comprising an updated amount of text capped at the new word limit, an updated amount of graphics capped at the media limit, and an updated amount of questions capped at the assessment limit;
• and
• re-render the updated course content on the UI in real time.

A non-transitory computer readable medium storing thereon computer executable instructions for updating a user interface displaying content related to a topic based on user comprehension of the topic, including instructions for:

• receiving, via the user interface (UI), a user selection of the topic;
• generating, using a machine learning algorithm, a course from reference materials describing a plurality of sub-topics associated with the topic, wherein generating the course includes:
• filtering the reference materials to select materials that meet a threshold quality level and a preferred difficulty level,
• executing an unsupervised machine learning model utilizing a graph-based or clustering algorithm to identify and extract selected sentences from the reference materials,
• capping the extracted selected sentences such that a total length of course content is bound to a word limit generated for the sub-topic, and generating a plurality of assessments that test comprehension of the plurality of sub-topics;
• outputting the course on the UI;
• monitoring, in real-time, user interaction with both a first subset of the assessments within the course on the UI and course content tested by the first subset of the assessments;
• determining whether the user interaction meets a comprehension criterion by generating an input vector comprising a plurality of attributes associated with the user interaction and feeding the input vector into a machine learning classifier, wherein the machine learning classifier applies a plurality of learned weights to the input vector to output a classification of whether the user interaction meets the comprehension criterion, the plurality of learned weights being optimized by training the machine learning classifier on a training dataset comprising historical input vectors mapped to corresponding output labels indicating whether historical user interactions satisfy the comprehension criterion;
• in response to determining that the user interaction does not meet the comprehension criterion;
• automatically updating the course content displayed in the UI by executing at least one machine learning model configured, individually or in combination, to:
• generate, for an input duration, a new word limit of text in the course content, a media limit of graphics in the course content, and an assessment limit of questions in the course content;
• and
• summarize the reference materials into the course content comprising an updated amount of text capped at the new word limit, an updated amount of graphics capped at the media limit, and an updated amount of questions capped at the assessment limit;
• and
• re-rendering the updated course content on the UI in real time.