This report contains: (i) A preliminary analysis of corpuses of text from blogs, news, and twitter; and (ii) An algorithm for a predictive text "R Shiny" app.

This slideshow explains the real estate clustering app found at 'https://jdallmann.shinyapps.io/comparablesapp/'.

This mock presentation consists in some preliminary investigation of variables influence on the price of homes in real estate data.

In this map, I use the latitude, longitude, and age of homes to cluster them into classes of natural comparables. Unlike the historical divisions that currently cut up real estate classes in any city, these are based on a natural classification to give insight into which homes are comparable.