In this project, we explore the historical data (from now until 6/27/2014) from Application Development Initiative's Density project: http://density.adicu.com/ Aspects of this project include:

• Module to scrape and explore Density's historical data in a readable format.
• Application of the above applied to Dining Hall and Library data - Dining Hall Occupancy Forecasting - For each semester, assume each day of the week has it's own unique pattern - Remove any days with less than 600-1000 total occupants (spring break, thanksgiving, etc) - Through User-GUI interaction, choose which dining hall and how many hrs. into future to forecast - Gather most up-to-date (<15min lag) data from Density API and train SVR classifier: - X ~ historical data of same timerange as up-to-date data, y ~ historical datapt directly after (15min) - Predict next point with X ~ current data - for multiple predictions, append prediction to X and repeat - Library Finals Season Analysis - Reccomendation engine for which library will have the least percentage occupancy.

Google Slide: https://docs.google.com/presentation/d/1y8WbdyCzO0mK8M26ygPWX7h4Y7I0EgMnDcF9n9hv0oE/edit?usp=sharing