Data Wrangling with dplyr - Part 3

dplyr_part_3.R

# install 
install.packages('dplyr')
install.packages('readr')

# library
library(dplyr)
library(readr)

# import data
ecom <- readr::read_csv('https://raw.githubusercontent.com/rsquaredacademy/datasets/master/web.csv')
ecom

# check the sources of traffic and device types.
ecom %>%
  distinct(referrer)

ecom %>%
  distinct(device)

# rename columns
ecom %>%
  rename(time_on_site = duration)

# sampling data
ecom %>%
  sample_n(700)

ecom %>%
  group_by(referrer) %>%
  sample_n(100)

ecom %>%
  sample_frac(size = 0.7)

ecom %>%
  group_by(referrer) %>%
  sample_frac(0.3)

# extract the `device` column.
ecom %>%
  pull(device)

# extract the first column.
ecom %>%
  pull(1)

# extract the last column
ecom %>%
  pull(-1)

# extract the first 20 rows
ecom %>%
  slice(1:20)

# extract the last row
ecom %>%
  slice(n())

# total number of observations in the data
ecom %>%
  tally()

# observations of different types of referrers
ecom %>%
  group_by(referrer) %>%
  tally()

# observations of referrers and bouncers
ecom %>%
  group_by(referrer, bouncers) %>%
  tally()

# observations of referrers and purchasers
ecom %>%
  group_by(referrer, purchase) %>%
  tally()

ecom %>%
  group_by(referrer, purchase) %>%
  tally() %>%
  filter(purchase == 'true')

# use `count()` instead of `tally()`
ecom %>%
  count(referrer, purchase)

# top 2 referrers that bring orders
ecom %>%
  count(referrer, purchase) %>%
  filter(purchase == 'true') %>%
  arrange(desc(n)) %>%
  top_n(n = 2)

ecom %>%
  pull(n_pages) %>%
  between(5, 15) 

mtcars %>%
  select(mpg, disp, cyl, gear, carb) %>%
  mutate(
    type = case_when(
      disp > 200 ~ 'High',
      cyl == 8 ~ 'Eight',
      TRUE ~ 'True'
    )
  )

ecom %>%
  pull(referrer) %>%
  nth(1)

ecom %>%
  pull(referrer) %>%
  nth(1000)

ecom %>%
  pull(referrer) %>%
  last()

ecom %>%
  pull(referrer) %>%
  first()