My most recent project at General Assembly utilized data from Spotify to predict song popularity and used a couple of statistical modeling methods. Through hands-on data acquirement, custom Python functions, and querying their external API I was able to collect around 16,000 songs and their respective audio features. Furthermore, by extensive research and perfecting functions, I wrangled those 16,000 songs and created a usable dataframe, in .csv format, ready for modeling. Before I began modeling, I dove into my dataset I created and performed exploratory analysis. This is one of my favorite parts of analysis because I can see the pre-existing underlying trends, discover patterns that I didnt expect, and research anomalies in the data. Once EDA and feature engineering was complete, I went onto modeling. I ran many applied statistical models including various regressions, decision trees, and neural networks. I then evaluated these models with their preferred metrics and presented some of the best ones. My organized repository on Github contains all of my code and a pdf of the presentation I gave.
I am looking to work for an established company where I can use my data analysis skills and also collaborate with other talented analysts. I would like to run lots of exploratory analysis and modeling. I am excited at the opportunity of having my analyses have an impact on business decisions. I'd like to be able to tackle new projects with a team and follow through beginning to end.
Work in United States
Expected Base Salary
Total Professional Experience
Send a connection request to the candidate to get their contact details.Contact Candidate