Aggressive characters in Harry Potter

harry_potter_aggression.R

#----------------
# Load packages
#----------------
library(RCurl)
library(dplyr)
library(tidyr)
library(ggplot2)
library(grid)
library(Cairo)


#-------------------
# Helper functions
#-------------------
# Nice clean, stripped-down theme
theme_clean <- function(base_size=12, base_family="Source Sans Pro Light") {
  ret <- theme_bw(base_size, base_family) + 
    theme(panel.background = element_rect(fill="#ffffff", colour=NA),
          axis.title.x=element_text(vjust=-0.2), axis.title.y=element_text(vjust=1.5),
          title=element_text(vjust=1.2, family="Source Sans Pro Semibold"),
          panel.border = element_blank(), axis.line=element_blank(),
          panel.grid=element_blank(), axis.ticks=element_blank(),
          legend.position="bottom", 
          axis.title=element_text(size=rel(0.8), family="Source Sans Pro Semibold"),
          strip.text=element_text(size=rel(1), family="Source Sans Pro Semibold"),
          strip.background=element_rect(fill="#ffffff", colour=NA),
          panel.margin.y=unit(1.5, "lines"))
  
  ret
}

# Append spaces to labels to fake right margin/padding
add.spaces <- function(x) {
  return(as.character(paste0(x, "   ")))
}


#------------
# Load data
#------------
# URL to the fantastic data
data.url <- "https://docs.google.com/spreadsheets/d/1heSMqYzYnL5bS0xiZ2waReUMLKIxHce5MMsQxzhgaA8/export?format=csv&gid=384008866"

# Book names
nice.books <- c("The Sorcerer's Stone", "The Chamber of Secrets", 
                "The Prisoner of Azkaban", "The Goblet of Fire", 
                "The Order of the Phoenix", "The Half-Blood Prince", 
                "The Deathly Hallows")

# Load and rearrange the data
hp <- read.csv(textConnection(getURL(data.url)), as.is=TRUE) %>% 
  select(1:13) %>%  # Only get the first 13 columns
  select(-c(g_e_m_n, evil, creature, tot)) %>%  # Get rid of other columns
  gather(book, aggressions, -c(Name, abb)) %>%
  mutate(book = factor(book, levels=levels(book), labels=nice.books, ordered=TRUE),
         book.rev = factor(book, levels=rev(levels(book)), ordered=TRUE))


#------------
# Plot data
#------------
# Find most aggressive characters
hp %>% group_by(Name) %>% summarize(Aggressions = sum(aggressions)) %>% 
  arrange(desc(Aggressions)) %>% head(5) -> most.aggressive

# Rearrange data for plotting
plot.data <- hp %>%
  filter(Name %in% most.aggressive$Name) %>%
  mutate(Name = factor(Name, levels=most.aggressive$Name, ordered=TRUE),
         Name.rev = factor(Name, levels=rev(levels(Name)), 
                           labels=add.spaces(rev(levels(Name))), ordered=TRUE)) %>%
  group_by(Name.rev, book.rev) %>%
  summarize(aggr = sum(aggressions))

# Plot top 5
ggplot(plot.data, aes(x=book.rev, y=aggr, fill=Name.rev)) + 
  geom_bar(stat="identity", position="dodge") + 
  geom_hline(yintercept=seq(10, 50, by=10), colour="#ffffff", size=0.25) + 
  coord_flip() + 
  labs(x=NULL, y="Instances of aggression", 
       title="Most aggressive characters in the Harry Potter series") + 
  scale_fill_manual(values=c("#915944", "#9BAB9C", "#692521", "#CAAA17", "#B51616"), 
                    guide=guide_legend(reverse=TRUE), name="") + 
  theme_clean() + theme(legend.key.size = unit(0.5, "lines")) -> nice.plot

# Save plot
ggsave(nice.plot, filename="top_5.png", width=7, height=5, units="in")
ggsave(nice.plot, filename="top_5.pdf", width=7, height=5, units="in", device=cairo_pdf)

how you got the csv data file? was it available or you created it? Anyways, awesome work!!