projects.mako.cc - state_of_wikimedia_research_2013/blob - figures/cite_graph.R

   1 # the last line is projected based on citations to the end of october
   2 # (almost certainly conservative)
   3 library(ggplot2)
   4
   5 d <- read.csv("wikipedia_citations.txt",header=F)
   6 colnames(d) <- c("year", "citations")
   7
   8 # print the total number of citations
   9 sum(d$citations)
  10
  11 # generate and print a graph
  12 # p <- qplot(year, citations, data=d) +
  13 #  geom_line(colour="blue") + geom_point(colour="blue")
  14
  15 p <- qplot(factor(year), citations, data=d, geom="bar", fill=I("darkblue"))
  16 p <- p + scale_x_discrete("Year") + scale_y_continuous("Number of Papers")
  17
  18 pdf("citations_by_year.pdf", width=7.5, height=5.3)
  19 print(p)
  20 dev.off()
  21
  22 ## data from dario
  23 ##########################################################3
  24
  25 # import data from dario
  26 d <- read.csv("Wikipedia publications - Data.csv")
  27
  28 # clean up the dates
  29 colnames(d)[1] <- "date"
  30 d <- d[,c(-9,-10)]
  31 d <- d[!d$date == "2013 to date",]
  32 d$date <- as.factor(d$date)
  33
  34 library(reshape)
  35 qplot(date, value, data=melt(d), group=variable, geom="line") +
  36   aes(colour=variable) + scale_y_log10()