Nicholas Dougherty

While working alongside three other Data Scientist, our team integrated open- source movie databases, examining the elements and attributes of film success, through implementing several varying data types. A classification model was the most apt approach for this problem set, and after I finished acquiring and preparing the data for our team, I ran repeated cross-validation efforts via Scikit-learn’s GridSearchCV; exhaustively searching over several variations of parameters values across every classification estimator applicable to our data frame. These efforts resulted in a successful, significant improvement over the baseline. After recording these success metrics in Jupyter Notebook, we shared our findings with our leaders and peers via live a presentation on Zoom.