rodneymawero / videos

Jupyter notebooks describing various data science analyses on supplementary online videos

The AIM - Analyzing supplementary videos

I lecture a first year engineering course and during the semester I created 80 supplementary videos that the students could watch in their own time. I was interested to see how the videos affected the final semester grades.

To do this, I utilized machine learning, specifically feature selection, to identify which of the 80 videos (features) had the strongest association with the target final semester grades.

Data

The data set is called FYMarksMinute2.csv. There are 80 features, each representing the watch time for a different video. The target is the final year grades for each student. And finally, there are 712 students (observations).

Coding

I run all the analyses in Python Jupyter Notebook. In this analysis I have made use of sklearn and various libraries. The Jupyter Notebook is called Videos feature selection.ipynb