From preliminary data wrangling with Python, the result shows an average of 2.3% completion rate for courses across 10 disciplines. The goal of this project is to help Canvas promotes course engagement, help students improve completion and motivate self-regulated learning.
Canvas Network open dataset (https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/1XORAL)
A webpage that displays academic progress of an online course, it darfts a report of the class average, the user's grade, completed percentage and the academic ranking of the user based on their grades.
Course recommendation uses collaborative filtering to predict how likely it is for the user to complete this new course based on the user’s past completed courses and what other people who are similar to the users have completed.
Frontend: HTML5, CSS, D3.js, Javascript
Backend: Flask, Python
Machine Learning: SciKit Surprise, pandas
Goto https://canvasprogress.herokuapp.com/, try entering "832312726" or "832670873" as the User ID (or any User ID from the dataset).