Group 2: Assignment 1

[rbjoern]

library("readr") library ("dplyr") library("stringr") library("ggplot2") library("lubridate") library("maps")

df = read_csv("https://raw.githubusercontent.com/MuseumofModernArt/collection/master/Artworks.csv")

Question 1

Create a new dataframe of the stock of paintings at

MOMA for each month in the year.

df$month = months(df$DateAcquired) Paintings <- data.frame(subset(df, Classification == "Painting")) Paintings_month <- data.frame(table (Paintings$Classification, Paintings$month)) Paintings_month$Var1 <- NULL

Question 2

Use ggplot2 and your new data frame

to plot the the stock of paintings on the y-axis and the date on the x-axis.

what kind of geom do you think is appropriate? why?

color the geom you have chosen red

add a title and custom axis labels

Paintings_month <- transform(Paintings_month, Var2 = reorder(Var2, Freq)) p = ggplot(data = Paintings_month, aes(x = Var2 , y = Freq)) p + geom_point(fill="red", colour="red") + ggtitle("Stock of paintings at MOMA") + labs(x="Month", y="Stock of pantings")

Question 3

Create the same plot but this time the color should reflect the stock of paintings for

curator approved and non-curator approved paintings, respectively

DataQ3 <- data.frame(table (Paintings$Classification, Paintings$month, Paintings$CuratorApproved)) DataQ3$Var1 <- NULL DataQ3 <- transform(DataQ3, Var2 = reorder(Var2, Freq)) colnames(DataQ3)[colnames(DataQ3)=="Var3"] <- "CuratorApproved" p = ggplot(DataQ3, aes(x=Var2, y=Freq, colour = CuratorApproved)) p + geom_point(stat = "identity") + coord_flip() p + geom_point(fill="blue") + ggtitle("Stock of paintings at MOMA") + labs(x="Month", y="Stock of pantings")

Question 4

Create a new data frame of the stock of paintings grouped by what department the painting belongs to.

Department_sum<- data.frame(count(Paintings, Department, wt=NULL, sort=FALSE))

Question 5

Plot this data frame using ggplot2. Which department has had

the highest increase in their stock of paintings?

Plotting & flipping

p = ggplot(Department_sum, aes(x = Department, y = n)) p + geom_bar(stat = "identity") + coord_flip()

Solution: As seen on Barplot the department with the highest increase in their stock of paintings is

Paintings And sculptures.

Question 6

Write a piece of code that counts the number of paintings by each artist in the data set.

List the 10 painters with the highest number of paintings in MOMAs collection.

Artists<- data.frame(count(Paintings, Artist, wt=NULL, sort=T))

Changing Missing name to Unknown Artist"

Artists[Artists == ""] <- "Unknown Artist"

Listing the 10 artists with the highest number of paintings.

head(Artists,n=10)

Solution: As seen in the console the artist with the highest number of paintings is Picasso, followed by Henri Matisse.

Question 7

The variable ArtistBio lists the birth place of each painter.

Use this information to create a world map where each country is colored according

to the stock of paintings in MOMAs collection.'

df.paintings <- subset(df, Classification == "Painting")

Extract nationality from ArtistBio

df.paintings$Demonym.1 <- str_extract(df.paintings$ArtistBio, "[A-z]*, ") df.paintings$Demonym.1 <- gsub(", ", "", df.paintings$Demonym.1)

Data: Count paintings for each nationality

df.country <- tally(group_by(df.paintings, Demonym.1))

We seek aid from a list of countries based on nationalities

countrylist <- read.csv("http://t2a.co/blog/wp-content/uploads/2014/03/Countries-List.csv")

... and merge this data with our own to gain country names

df.countrycoded <- merge(df.country, countrylist, by = "Demonym.1" ) df.countrycoded <- subset(df.countrycoded, df.countrycoded$Demonym.1 != "") # We ignore the less welldefined ones df.countrycoded <- subset(df.countrycoded, df.countrycoded$Name != "Congo (Republic of)") # We do not wish to count the Congolese gentlemen for both Congos

We lose twenty paintings without welldefined countries, which is deemed acceptable

sum(df.country$n) # 2229 sum(df.countrycoded$n) # 229

We now begin construction of the map!

map <- map_data("world")

First, we find out how well the country names in our data fit the maps package

map.1 <- left_join(df.countrycoded, map, by = c("Name" = "region"))

subset(map.1, is.na(map.1$long)) # Oh no! We cant find three of our countries

Manual correction (sorry)

df.countrycoded$Name <- gsub("United States of America", "USA", df.countrycoded$Name) df.countrycoded$Name <- gsub("United Kingdom", "UK", df.countrycoded$Name) df.countrycoded$Name <- gsub("Russian Federation", "Russia", df.countrycoded$Name)

Recheck

map.1 <- left_join(df.countrycoded, map, by = c("Name" = "region")) subset(map.1, is.na(map.1$long)) # Found them!

We use a right join instead, so as to make countries without paintings grey rather than removing them.

Now, they merely have NA in the n variable

map.2 <- right_join(df.countrycoded, map, by = c("Name" = "region"))

We then construct the world map

p = ggplot(map.2, aes(x = long, y = lat, group = group, fill = n)) p + geom_polygon() + expand_limits(x = map.2$long, y = map.2$lat)

Question 8

The Dimensions variable lists the dimensions of each painting.

Use your data manipulation skills to calculate the area of each painting (in cm).

Create a data frame of the five largest and five smallest paintings in MOMAs collection.

Creating new variabel which include the text in parenthesis

Paintings$cm = gsub(".((.)).*", "\1", Paintings$Dimensions)

Taking the height fromt the cm variable

Paintings$height_cm = gsub( " .*$", "", Paintings$cm)

Finding the lenght from the cm variable

Paintings$length_cm = str_extract(Paintings$cm, "x [0-9]*.[0-9] cm") Paintings$length_cm = gsub("x ", "", Paintings$length_cm) Paintings$length_cm = gsub(" cm", "", Paintings$length_cm)

Creating variable for size

Paintings$size = (as.numeric(Paintings$height_cm))*(as.numeric(Paintings$length_cm))

Sort desc

Paintings = Paintings[order(-Paintings$size),]

Creating new dataset with highest 5 sizes

top = head(Paintings,5)

Sort asc

Paintings = Paintings[order(Paintings$size),]

Creating new dataset with smallest 5 sizes

bottom = head(Paintings,5)

Merge two new dataframes

Together = merge(top, bottom, all=TRUE)

Solution: The dataset Togehter include the five smallest and five largest paintings.

[sebastianbarfort]

Generally, I think you should use geom_line instead of geom_point as you're plotting the same variable over time.

I can't make this line work:

countrylist <- read.csv("http://t2a.co/blog/wp-content/uploads/2014/03/Countries-List.csv")

Can you please check and update the code above when you're sure it works.

[rbjoern]

Hi Sebastian.

The code works fine for me (i was also the one who checked it prior to submitting). When i simply run the entire code, a working map appears, which can only happen if this particular line works as well. The csv file is also loaded correctly as a dataframe. Furthermore, when i open the link in a browser, the csv file opens correctly.

[sebastianbarfort]

Generally a pretty good assignment.

I would recommend using the functions from the dplyr package more systematically throughout your code. I find it makes it much easier to do data manipulation tasks. For example, arrange is easier to use than order, filter is easier than subset and so on.

APPROVED