Introduction
Hello, my name is Kevin Ryan, and I am from southern New Jersey, right outside Philadelphia, PA. My journey in programming, data analysis, visualization, etc. started in August 2020 when I began my Master’s program in Buisness Analytics at the University of Notre Dame in Notre Dame, Indiana. My expected graduation date will be May 2021.
I will be constantly updating this repository as I move throughout the program to highlight some of the completed coursework and projects that I feel confident in enough to publish. I hope all enjoy.
Projects
Machine Learning - Spotify and Popularity
- As part of a group of 3, I analyzed a Spotify dataset from Kaggle and used principles learned in my Machine Learning class to study what characteristics and what levels of those characteristics contributed to songs’ popularity ratings. Using an XGBoost Model and SHAP variable importance to reach our conclusions, we found that more recent songs are most popular with Spotify listeners, with most of those songs exhibiting low levels of acoustics, less lyrics, and high levels of “danceability”.
- Dataset: https://www.kaggle.com/yamaerenay/spotify-dataset-19212020-160k-tracks
- Easy access to HTML: https://htmlpreview.github.io/?https://github.com/fkryan73/ND_MSBA/blob/main/spotify_analysis_final.html