Looks at "mt vernon va" search frequency over time to determine most popular months for visiting Washington's Mt. Vernon estate

mt_vernon.r

# Looks at "mt vernon va" search frequency over time to determine most popular
# months for visiting Washington's Mt. Vernon estate

library(rCharts)
library(dplyr)
library(stringr)
library(RCurl)

# Read in the data
x <- getURL("https://gist.githubusercontent.com/AmritPatel/4665ebd02f810fc10b37/raw/b24e1c6210e21c89db2aa4037b540bfebf26cda0/mt_vernon.csv")
mt_vernon <- read.csv(text = x)

# Use dataframe wrapper
mt_vernon_df <- tbl_df(mt_vernon)

# Break up the date filed by month, day, and year for futher processing
break_date <- mutate(mt_vernon_df,
                     month = str_sub(str_extract(Date, "\\d\\d?/"),             end = -2),
                     day   = str_sub(str_extract(Date, "/\\d\\d?/"), start = 2, end = -2),
                     year  = str_sub(str_extract(Date, "/\\d\\d?$"),  start = 2) 
                     )

# Only select the columns we need
case <- select(break_date, month, year, mt.vernon.va)

# Plot of data as a function of year (grouped by month)
r1 <- nPlot(mt.vernon.va ~ year,
            group = "month",
            data = case,
            type = 'multiBarChart',
            size = list(const = 3)
            )

# Average the monthly data over years
month_avg <- summarise(group_by(case, month),
                       avg_over_years = mean(mt.vernon.va)
                       )

# Plot the monthly data average over years
r2 <-  nPlot(avg_over_years ~ month,
             data = month_avg,
             type = 'multiBarChart',
             size = list(const = 3)
             )
# Chart available @ http://rcharts.github.io/viewer/?c24af1f90fd5f6370bfc           
#r2$publish("\"mt vernon va\" search frequency over time to determine most popular months for visiting Washington's Mt. Vernon estate", host = 'gist')