<img height="1" width="1" style="display:none" src="https://www.facebook.com/tr?id=363869954087733&amp;ev=PageView&amp;noscript=1">
Adaptive eLearning: How to Leverage Content For a Personalized Study Plan Blog Feature

Adaptive eLearning: How to Leverage Content For a Personalized Study Plan

In Part 1 of our two-part series on BenchPrep’s adaptive learning technology, I explained how we rank our content based on importance and difficulty using the data we collect using a process we call Learning Engine Optimization (LEO).

In Part 2, I will explain how we take that content and actually create the BenchPrep Adaptive Study Plan using even more data. The Adaptive Study Plan means that the Learning Engine modifies the presentation of material in response to student performance—something that a 400-page bound book might have trouble doing. I have tried to make it more readable by not using a technical terms. I will share the technical details of our Adaptive system, as well. Let’s jump in!

Step 1. Scope the content based on the selected course.

Right now, we build adaptive study plans at the course level, and most of these courses prepare students for standardized tests. For example, two users in our GRE Prep by McGraw-Hill course will have a study plan based on the same content, but the order and amount of content they each consume might be different, based on, for example, when they plan to take the test. This seems easy enough.

Step 2. Order the content files based on their Learning Engine Optimization (LEO) rank.

Once we’ve identified all of the files in a course, we use the rankings from LEO to order the content files. The initial ordering is based on a file’s importance and difficulty, which is relative to each test. By starting with the easier material, we can better gauge how to adapt the material as users progress through the course.

Step 3. Combine ordered learning modules with student performance.

This is where things get tricky. The purpose of an adaptive study plan is that it’s personalized. Each user will start with the same easy and important pieces of content. After that, things change. Users are fed different pieces of content depending on their performance on previous files. For example, if one of our users in the GRE Prep by McGraw-Hill course answered all of the geometry practice questions correctly, he will move on to the next important topic. However, if the another user in the GRE Prep by McGraw Hill course only answered a few of them correctly, he will be prompted to read/watch lessons and attempt more geometry practice problems before moving to the next topic.

We use the strengths and weaknesses of each user and machine-learning techniques to determine the next piece of content that suits them best.

We're trying to create personalization— on a massive scale. No two students are exactly alike, nor are two test questions or two topics. By reimagining and redesigning traditional course content as dynamic, discrete, measurable elements of an adaptive learning system, we have laid the foundation for accelerated learning from the user end— and for accelerated learning on our end too, about how others learn!

As you can see, a lot happens behind the screen, and it’s our responsibility to share. Hopefully these blog posts will help you think the same thing too. Not only will learners improve their test scores and learn new material, but they will do it faster than ever.

Curious to know how McGraw-Hill Education leveraged BenchPrep's digital learning platform to maintain their established position as an industry leader? Download our case study, Taking Traditional Learning Methods Into The Digital Stratosphere.

 Download Now 

 

About Ujjwal Gupta

Chief Operating Officer & Cofounder at BenchPrep

  • Connect with Ujjwal Gupta
eLearning Learning