Analysis of Gmail using the R language.

mail.R

library(ggplot2)
library(scales)

data <- read.table("index.csv", header=FALSE, sep=",", quote="\"")
dim(data)

title <- data[,c("V1")]
from  <- data[,c("V2")]
to    <- data[,c("V3")]
date  <- data[,c("V4")]


#--from--#

length(from)
head(from)

from  <- gsub(".*<|>.*", "", from)
tail(from)

from.table <- table(from)
from <- as.data.frame(from.table)
from$per <- (from$Freq/sum(from$Freq))*100

summary(from$Freq)
from.sort <- order(-from$Freq)
from <- from[from.sort,]
head(from, n=20)

#--END--#

date.tmp <- date

#--date--#

date <- gsub(" .*", "", date)
date.table <- table(date)
date <- as.data.frame(date.table)
date$date = as.POSIXct(date$date, format="%Y/%m/%d")
head(date)

ggplot(date, aes(x=date, y=Freq, group=1)) + geom_line() +
  scale_x_datetime(labels=date_format("%Y-%m-%d")) +
  ggtitle("受信したメールの受信数推移") +
  xlab("日付") + ylab("受信数")
  
#--END--#


#--time--#

time  <- gsub(".* ", "", date.tmp)
time.table <- table(time)
time <- as.data.frame(time.table)
time$time = as.POSIXct(time$time, format="%H:%M")

head(time)
ggplot(time, aes(x=time, y=Freq, group=1)) + geom_line() +
  scale_x_datetime(labels=date_format("%H:%M")) +
  ggtitle("受信したメールの時間別グラフ") +
  xlab("時間") + ylab("受信数")
  
#--END--#