This classifier uses app store text reviews to predict which app category the review belonged to. I learned how to separate insight from noise by evaluating multiple metrics from different angles. This project was implemented in R, and my ensemble model attained an accuracy of 0.61.