This report analyzes national flags, using 193 countries flags to see if flag colors are associated with certain regions of the world. This report goes into detail about image processing, color identification, and categorical methods of testing the relationship between colors and global areas.

Project 8, April 2021. Used Linear Discriminant Analysis and Logistic Regression to evaluate if bank customers were more likely to be loan acceptors or decliners.

Project #6, March 2021. Used the rpart package to create trees and prune them, examining error rates and outliers, regarding the price of Toyota Corollas

Project #7, March 2021. Used the neuralnet package to generate neural networks with various number of layers and nodes. Also used RMSE to compare models and reisudal plot to examine potential outliers.

Project #5, March 2021. Used confusion matrices and lift charts to calculate model error rates.

Project #4, March 2021. Used regularization methods: ridge, lasso, and elastic net to model housing in the Boston area.

Project #3 for Data Mining. February 2021. This project uses PCA to reduce the number of x variables in our model. Scree plots, biplots, and correlation matrices were used prior to fitting the linear regression model.

Project #2 for Data Mining STAT 415. February 2021. This project helped me better understand some of the basics of text mining.

Project #1 for Data Mining STAT 415. January 2021. This project explores and summarizes some of the responses Kaggle received in their 2020 survey.

Using a probability transition matrix to generate 10,000 transitions of a markov chain to determine long-term probability for the # of users connected.

This project uses the Acceptance-Rejection Method to generate random samples for X when the pdf is given, but the cdf is unknown and/or complicated.

Using Monte Carlo Simulation to simulate long-term probability

This is the documentation for the first version of the Shiny App I am creating. It uses Leaflet to geographically visualize the data in The Washington Post's 'fatal police shootings' dataset.