Project Two

Maks Nikiforov and Mark Austin

Purpose

This repository contains files needed for analysis of an Online News Popularity data set. renderProjectTwo.R contains render code that works with input file ProjectTwo.Rmd to produce different github document files for each of the 6 channels. Data summaries and modeling are presented for each channel.

Github pages is enabled to allow viewing the channel analysis pages as linked html files.

Required R Packages

The following R packages are required to run R code used in this project and/or create the linked channel documents.

• tidyverse The tidyverse package is used for data handling and plotting.

• caret The caret package is used for working with data and models.

• corrplot The corrplot package is used to produce a visual correlation plot.

• knitr The knitr package is used for document image handling.

• rmarkdown The rmarkdown package is used by a render program to render this document.

• doParallel The doParallel package is used to enable parallel processing with the caret package

• lares The lares package provides the corr_cross() function to compute and return the highest and most significant variable correlations in a data frame.

• rvest The rvest package provides the functions read_html() and html_text() to scrape the titles of top ten Mashable articles in each category of the online news popularity data set.

Analyses by Channel

Business articles is available here

Entertainment articles is available here

Lifestyle articles is available here

Social Media articles is available here

Tech articles is available here

World articles is available here

Render Code Used to Create Analyses

library(tidyverse)
library(rmarkdown)

##Read data then extract data channel variables
fullData<-read_csv("./data/OnlineNewsPopularity.csv")

##Extract data channel variables from column names
channels<-names(fullData)[startsWith(names(fullData),"data_channel")]
channels

##Create .md filenames
output_file <- paste0(channels, ".md")
output_file

#create a list for each channel with just the channel name parameter
params = lapply(channels, FUN = function(x){list(channel = x)})


#put into a data frame 
reports <- tibble(output_file, params)
reports

#use render to create report for each channel
# by passing channel parameter
apply(reports, MARGIN = 1,
      FUN = function(x){
        render(input = "ProjectTwo.Rmd",
               output_format = "github_document",
               output_file = x[[1]],
               output_options = list(
                 html_preview = FALSE,
                 df_print = "tibble",
                 toc = TRUE
               ),
               params = x[[2]])
      })