Support Vector Machines

EDA

It's always a good idea to perform an exploratory data analysis (EDA) before doing any type of predictive modeling. In this stage of the project I was looking for correlation between variables and any other noteworthy characteristic which would have an impact on the results of our modeling.

SVMs

I tested and compared 4 different support vector machine predictive models and found the results to be desirable. With this being the case, I did find a drawback of SVMs which is their long computation times.

Machine Learning for Business

For this project I explored a Spotify song dataset which contained descriptive variables that measure a song's popularity, energy, and more. I proposed a hypothetical business situation where we are tasked with building the most appropriate playlist for our client's bar in order to attract the most customers.