A Jupyter notebook that uses a dataset of over 5000+ movies scraped from IMDb, and attempts to predict whether a movie will be highly rated based on the training data.
This notebook is written in Python and uses Pandas for transforming and cleaning the dataset, Matplotlib for visulization, and Scikit-learn for machine learning.
Using a genetic algorithm, and real data from Google Maps, we try to find the quickest route that visits all 49 capitals in the Continental United States.
This project is written in Python, and uses several different Google Map APIs: Directions, Distance Matrix, Static Maps, and Geocoding. The genetic algorithm implements stochastic selection and partionally matched crossover.