erikgregorywebb

  # calculate character % present
  character_importance_plot = all_characters %>% 
    mutate(show = factor(show, levels = c('The Office', 'Parks & Recreation', 'Modern Family', 'Community', 'New Girl', 'The Good Place'))) %>%
    mutate(per = character_episodes / total_episodes) %>%
    ggplot(., aes(x = reorder(character_name , per), y = per, col = show)) + 
    geom_point(size = 3) + 
    coord_flip() + 
    theme(legend.position = 'none') +
    labs(title = 'Character Importance by Show', subtitle = 'IMDb Episode Descriptions, All Seasons',
         y = 'Percent Present in Episode Descriptions', x = '', fill = 'Season', caption = 'Data Source: IDMb.com | Author: @erikgregorywebb') +

erikgregorywebb

  # define character count function
  get_character_counts = function(df, show_name, character_name) {
    total_episodes = df %>% filter(show == show_name) %>% nrow()
    character_episodes = df %>% 
      filter(show == show_name) %>% filter(str_detect(description, character_name)) %>% nrow()
    result = tibble(show = show_name, character_name = character_name, 
                    total_episodes = total_episodes, character_episodes = character_episodes)
    return(result)
  }

erikgregorywebb

  rating_trends_plot = all_shows %>%
    mutate(season = factor(season, levels = c(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11))) %>%
    mutate(show = factor(show, levels = c('The Office', 'Parks & Recreation', 'Modern Family', 'Community', 'New Girl', 'The Good Place'))) %>%
    ggplot(., aes(x = airdate, y = rating, col = show, group = season)) + 
    geom_line(alpha = .5) + geom_smooth(se = F, size = 2) +
    theme(legend.position = 'none') +
    labs(title = 'Ratings Trend by Show', subtitle = 'IMDb Episode Average Rating, All Seasons',
         y = 'Average Rating', x = 'Year', fill = 'Season', caption = 'Data Source: IDMb.com | Author: @erikgregorywebb') +
    facet_wrap(~show, scales = 'free') +
    scale_y_continuous(limits = c(6, 10)) + 

erikgregorywebb

  # define scraper function
  scrape_show = function(name, imdb_id, no_seasons) {
    datalist = list()
    n = 1
    for (i in 1:no_seasons) {
      Sys.sleep(3)
      url = paste('https://www.imdb.com/title/', imdb_id, '/episodes?season=', i , sep = '')
      page = read_html(url)
      episodes = page %>% html_node('.list') %>% html_nodes('.info')
      for (j in 1:length(episodes)) {

erikgregorywebb

  # define list of shows
  shows = tibble(
    name = c('The Office', 'Parks & Recreation', 'Modern Family', 'Community', 'New Girl', 'The Good Place'),
    imdb_id = c('tt0386676', 'tt1266020', 'tt1442437', 'tt1439629', 'tt1826940', 'tt4955642'),
    no_seasons = c(9, 7, 11, 6, 7, 4),
  )

erikgregorywebb

  # import packages
  library(tidyverse)
  library(rvest)
  library(lubridate)
  library(scales)

erikgregorywebb

# import packages
library(tidyverse)
library(rvest)
library(lubridate)
library(scales)

# define list of shows
shows = tibble(
  name = c('The Office', 'Parks & Recreation', 'Modern Family', 'Community', 'New Girl', 'The Good Place'),
  imdb_id = c('tt0386676', 'tt1266020', 'tt1442437', 'tt1439629', 'tt1826940', 'tt4955642'),

erikgregorywebb

# import
library(rJava)
library(rMouse)
Sys.setenv(JAVA_HOME='C:/Program Files/Java/jre1.8.0_241') # for 64-bit version

# loop
condition = FALSE
while(condition == FALSE) {
  Sys.sleep(30)

erikgregorywebb

library(rvest)
library(stringr)

# get list of gists
all_links = c()
for (i in 1:25) {
  url = paste('https://gist.github.com/erikgregorywebb?page=', i, sep ='')
  print(url)
  Sys.sleep(1)
  page = read_html(url)

erikgregorywebb

library(tidyverse)
library(rvest)
library(tools)
library(fuzzyjoin)

# extract list of programs by state (source: mastersindatascience.org)
url = 'https://www.mastersindatascience.org/schools/'
page = read_html(url)
states = page %>% html_nodes('.row') %>% html_nodes('a') %>% html_attr('href')