Currently, the implementation phase is at the stage of data collection, organization, and preprocessing before using ML tools (see figure 1). Python and various ML libraries are used to showcase the results of these implementations.
There are various Python implementations for the MF including Non-negative factorization (NMF), orthogonal factorization (SVD), and probabilistic factorization (PMF).
A comparative analysis is conducted to show which strategy and techniques that would be best to improve student learning experience and engagement. This approach will attempt to use resources such as student grade reports, LMS, e-text platforms, and other available sources to make meaningful and realistic suggestions to each student that allow her/him to see good progress.
The system should generate personalized learning plans and use suggestions from students who were successful in particular areas (courses) and be trustworthy enough to provide recommendations to "at risk" students.
So, the proposed system will also use other information such the login time students used in a particular content and crosscheck the corresponding grades, if available.
The analytics embedded in the system will have the option to systematically validate the recommendations based on students' performances. Students seeking help would tend to accept recommendations from their fellow students especially when coming those who have performed well in the same or similar courses.
This approach will give the opportunity to high achievers to not only provide feedback but also influence students with recommendations and enable them to improve their skills and performances.
For more details, please contact me here.