Inside Airbnb - Antwerp

run_antwerp.R

# Load packages
library(tidyverse)
library(reshape2)

# DOWNLOAD DATA 

## Function to download data and save as CSV
download_data <- function(url, filename){
  download.file(url = url, destfile = paste0(filename, ".csv"))
}

url_listings <- "http://data.insideairbnb.com/belgium/vlg/antwerp/2021-02-25/visualisations/listings.csv"
url_reviews <- "http://data.insideairbnb.com/belgium/vlg/antwerp/2021-02-25/visualisations/reviews.csv"

download_data(url_listings, "listings")
download_data(url_reviews, "reviews")

# CLEAN DATA 

reviews <- read_csv("reviews.csv")
listings <- read_csv("listings.csv")

## filter for reviews published since 01/01/2015
reviews_filtered <- reviews %>% filter(date > "2015-01-01")

## filter for `listings` that have received at least 1 review.
listings_filtered <- listings %>% filter(number_of_reviews > 1)

## merge the `reviews` and `listings` dataframes on a common columns (the type of join doesn't really matter since we already filtered out listings without any reviews)
df_merged <- reviews_filtered %>% inner_join(listings_filtered, by = c("listing_id" = "id"))

## group the number of reviews by month and neighborhood.
df_grouped <- df_merged %>%
  mutate(month = format(date, "%m"), year = format(date, "%Y")) %>%
  group_by(year, month, neighbourhood) %>%
  summarise(num_reviews = n())

## create date column
df_grouped <- df_grouped %>% mutate(date = as.Date(paste0(year, "-", month, "-01")))

## store the final data frame in `gen/data-preparation` as `aggregated_df.csv`
write_csv(df_grouped, "aggregated_df.csv")


# CREATE PIVOT TABLE

## import the data from `gen/data-preparation/aggregated_df.csv`
df <- read_csv("aggregated_df.csv")

## create pivot table 
df_pivot <- df %>% reshape2::dcast(date ~ neighbourhood, fun.aggregate = sum, value.var = "num_reviews")

## export results
write_csv(df_pivot, "pivot_table.csv")


# PLOT ANTWERP 

## import the data from `gen/analysis/pivot_table`
df_pivot <- read_csv("pivot_table.csv")

pdf("plot_Antwerp.pdf")
plot(x = df_pivot$date, 
     y = df_pivot$Universiteitsbuurt, 
     col = "red", 
     type = "l", 
     xlab = "",
     ylab = "Total number of reviews", 
     main = "Effect of COVID-19 pandemic on Airbnb review count")


lines(df_pivot$date, df_pivot$`Sint-Andries`, col="blue")
lines(df_pivot$date, df_pivot$`Centraal Station`, col="green")

legend("topleft", c("Universiteitsbuurt", "Sint Andries", "Centraal Station"), fill=c("red", "blue", "green"))
dev.off()

# PLOT ALL

## import the data from `gen/data-preparation/aggregated_df.csv`
df <- read_csv("aggregated_df.csv")

## group by date and calculate the sum of all reviews across neighbourhoods.
df_groupby <- df %>% group_by(date) %>% summarise(num_reviews = sum(num_reviews))

## plot the chart and store the visualisation.
pdf("plot_all.pdf")
plot(x = df_groupby$date, 
     y = df_groupby$num_reviews, 
     type = "l", 
     xlab = "",
     ylab = "Total number of reviews", 
     main = "Effect of COVID-19 pandemic on Airbnb review count")
dev.off()