This project was a result of a situation I encountered during my initial experience working in Learning & Development. I discovered that there was a vulnerability in the way the Learning Management System (LMS) we used at the time processed SCORM file packages, which created an opportunity for a low-tech exploit of the system such that learners could very easily trick the system into giving them the correct answer, while allowing them to retake the question without their score reflecting the initial missed question. The resulting grade inflation was only caught upon manual review of the SCORM reports. 

The solution I developed is a Python-based machine learning process that can analyze SCORM reports to identify the pattern caused by the exploit and automatically rescores the modules to show the earned grade of the learners rather than the exploit-inflated score. You can download the Python code from GitHub to analyze your own SCORM data and check whether your learners are using the exploit.

The project additionally includes some examples of how I can use Python for data analytics. I wrote a Python script to generate synthetic data to use for the project, so there are questions banks, an employee information database, SCORM reports based on those employees completing the assessments the question banks were created for, etc. to provide a realistic context for the project.