Art from Art

I was really excited to see that the Tidy Tuesday for last week (1/17) was on Art! I haven’t found a lot of great art data sets, and definitely not a lot with numerical data, so I was pretty interested in taking this data for a spin. The data is from the package arthistory, and was originally collected as part of an undergraduate thesis on demographic representation in Art History textbooks. I decided that rather than analyzing the set of data, I was more interested in how the data could be used to produce generative art. With that goal in mind, I dug into exploring the data.

Libraries and all that jazz

# install.packages("arthistory")
library(arthistory)

# install.packages("MetBrewer")

# Art themed art must be art colored
library(MetBrewer)

# essentially my utilities 
library(tidyverse)
library(here)
library(glue)

# setting the all important seed
set.seed(9902)
# from https://www.calculatorsoup.com/calculators/statistics/random-number-generator.php
# range 1 - 10,000

Initial Exploration

glimpse(worksjanson)

Rows: 1,634
Columns: 25
$ artist_name                      <chr> "A. R. Penck", "A. R. Penck", "Aaron …
$ artist_unique_id                 <dbl> 1, 1, 2, 2, 2, 2, 3, 3, 4, 4, 5, 6, 6…
$ artist_nationality               <chr> "German", "German", "American", "Amer…
$ artist_gender                    <chr> "Male", "Male", "Male", "Male", "Male…
$ artist_race                      <chr> "White", "White", "White", "White", "…
$ artist_ethnicity                 <chr> "Not Hispanic or Latinx", "Not Hispan…
$ title_of_work                    <chr> "The Demon of Curiosity", "The Demon …
$ edition_number                   <dbl> 5, 6, 3, 4, 5, 6, 5, 6, 7, 8, 7, 7, 8…
$ publication_year                 <dbl> 1995, 2001, 1986, 1991, 1995, 2001, 1…
$ part_in_text                     <dbl> 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4…
$ chapter_title                    <chr> "Post-Modernism", "Post-Modernism", "…
$ start_date                       <dbl> 1982, 1982, 1951, 1951, 1951, 1951, 1…
$ end_date                         <dbl> 1982, 1982, 1951, 1951, 1951, 1951, 1…
$ circa                            <chr> "0", "0", "0", "0", "0", "0", "0", "0…
$ medium                           <chr> "Acrylic on canvas", "Acrylic on canv…
$ height_of_work_in_book           <dbl> 8.5725, 8.5725, 11.1125, 11.1125, 10.…
$ width_of_work_in_book            <dbl> 8.8900, 8.8900, 8.8900, 8.8900, 8.890…
$ height_of_text                   <dbl> 12.0650, 10.7950, 3.1750, 3.1750, 3.1…
$ width_of_text                    <dbl> 8.890, 8.890, 8.890, 8.890, 8.890, 8.…
$ area_actual_work                 <chr> "78419.60123", "78419.60123", "0", "0…
$ area_of_work_in_book             <dbl> 76.20952, 76.20952, 98.79013, 98.7901…
$ area_of_text                     <dbl> 107.25785, 95.96755, 28.22575, 28.225…
$ work_to_figure_ratio             <chr> "1029", "1029", "0", "0", "0", "0", "…
$ work_in_text_in_color            <dbl> 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1…
$ location_as_catalogued_in_janson <chr> "Riverdell Collection, New York. Cour…

I was interested in using medium as the color for my art, however, it looks like there are a lot of different mediums listed, let’s take a look at all of them:

worksjanson |> 
  select(medium) |> 
  group_by(medium) |> 
  summarize(count = n()) |> 
  arrange(-count)

# A tibble: 168 × 2
   medium               count
   <chr>                <int>
 1 Oil on canvas          996
 2 Photograph              72
 3 Gelatin-silver print    64
 4 Acrylic on canvas       28
 5 Lithograph              27
 6 Oil on panel            18
 7 Collage                 15
 8 Albumen print           12
 9 Platinum print          11
10 Woodcut                 11
# ℹ 158 more rows

hi hi moo