DATA605 - Comp. Mathematics - Assignment 15 - Integration by U Substitution - Derivatives and 2nd Derivatives to find maximums, minimums, and saddle points

Re-upload due to update existing errors. Included an image

A comparison and exploration of SVMs for classifying a win vs Decision Trees.

A re-upload of the first since there were issues with updating the published doc Highlights some of the issues of decision trees and how random forests attempt to handle it.

library(knitr) library(data.table) library(readr) library(dplyr) library(randomForest) library(caret) library(ggplot2) library(tree) library(cowplot) library(corrplot) library(rmarkdown) Creating decision trees to predict a win with Nature's Prophet in Dota 2

Eigenvectors used on Shoe Images to show how linear algebra is used in Image Recognition with Eigenvectors from covariance matrices generated by PCA.

This assignment covers: 1. proving that matrix transposition is not equal to simple matrix multiplication 2. Using R code to perform LU Decomposition

An analysis on Self Control Capacity impacting Daily Habit Execution with data sourced by the following study A Longitudinal Field Study on the Role of Self-Control in Habit Formation [https://www.frontiersin.org/articles/10.3389/fpsyg.2020.00560/full] Techniques: - pivot_longer - aov (anova) - dplyr - tidyverse

This is the workspace I used to help build the app I have currently hosted at: https://hhl7f9-daniel-craig.shinyapps.io/Predicting_Wins_Classification/

A vignette extending classmate's work on Purrr

Classifying a "win" with hero Nature's Prophet from Dota 2 using variable from games. Trained on a dataset collected by OpenDota

A project focused on attempting to find if there is a higher success rate amongst those with self control at building new habits. Data and original study: https://www.frontiersin.org/articles/10.3389/fpsyg.2020.00560/full

A vignette for readr highlighting column specifications and column type setting during import to avoid tidying in the middle of a user's workflow down the line.

Works and cites through Text Mining with R by Silge and Robinson (2017)

This exercise uses an endpoint from NY Times API to pull JSON data, convert it to a dataframe, and then combine dataframes stored across the JSON/API endpoints from multiple calls into one database that would fit well into an SQL table and further analysis.

Pulling some JSON structures from Nobel's API (they have a super easy tutorial on their site and literally just give you the code.) Answering some questions after chopping and slicing some data up.

Focuses on importing data from different formats that you may receive when using an API. Some packages used here: library(rvest) library(htmltools) library(jsonlite) library(XML) library(RCurl) library(methods) Commands: html_element() html_table() jsonlite:: fromJSON() getURL() - used for getting the XML file xmlToDataFrame()

Data tidying practice using Kory's Therapists Untidy Data example from Week 5

Data tidying practice using Joe's Train Casualty Untidy Data example from Week 5

Data tidying practice using Jacob's NBA Rankings Untidy Data example from Week 5

This assignment focuses on the import of an untidy CSV file about airline arrival delays and formatting it into a more usable style.

Strong use of: - separate_wider_delim - some regex with replace to remove some formatting - apply() to apply functions across rows - basic data structure slicing - merge() It is mostly dynamic with only one section that doesn't respond to how many rounds of chess there are. 95% of this is dynamic to how many rounds of chess were played and can be performed on almost equally formatted file expect for the section titled 'lookup table prep.'

Primary Polls Subsetting + GGPlot