Skip to content

Commit 28545b0

Browse files
jilltxtjilltxt
committed
Initial commit
1 parent 8f5a430 commit 28545b0

File tree

1 file changed

+241
-0
lines changed

1 file changed

+241
-0
lines changed

‎machinevision_plots_by_year_country.R‎

Lines changed: 241 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,241 @@
1+
# R scripts to generate figures for the paper "Representations of Machine Vision
2+
# Technologies in Artworks, Games and Narratives: A Dataset", submitted to the
3+
# journal Data in Brief in March 2022.
4+
#
5+
# The code was written by Jill Walker Rettberg.
6+
#
7+
#
8+
9+
# setup------------------------------------------------
10+
11+
library(tidyverse)
12+
library(ggthemes)
13+
14+
# Import creativeworks ----------------------------------------------------
15+
16+
CreativeWorks <- read_csv("data/creativeworks.csv",
17+
col_types = cols(
18+
WorkID = col_integer(),
19+
WorkTitle = col_character(),
20+
Sentiment = col_factor(levels = c(
21+
"Exciting", "Flawed", "Helpful", "Neutral", "Wondrous",
22+
"Hostile","Oppressive", "Alien", "Creepy", "Subversive",
23+
"Dangerous", "Intrusive", "Empowering", "Protective",
24+
"Intimate", "Misleading", "Fun", "Overwhelming",
25+
"Prosocial", "Disgusting")),
26+
Topic = col_factor(levels = c(
27+
"Nudity", "Social Media", "Romantic relationship", "Climate Change",
28+
"Dystopian", "Horror", "Robots/androids", "Surveillance", "Automation",
29+
"City", "Labour", "War", "Identity", "AI", "Animals", "Consciousness",
30+
"Nature", "Companionship", "Competition", "Playful", "Family",
31+
"Free will", "Physical violence", "Crime", "Hacking", "Conflict",
32+
"Empathy", "Utopian", "Race", "Sex", "Cyborgs", "Inequality",
33+
"Economy", "Grief", "Autonomous vehicles", "Gender")),
34+
TechRef= col_factor(levels = c(
35+
"Holograms", "Augmented reality", "Ocular implant",
36+
"Emotion recognition", "Surveillance cameras", "AI",
37+
"Virtual reality", "Motion tracking", "Body scans",
38+
"Drones", "MicroscopeOrTelescope", "Biometrics",
39+
"Image generation", "Facial recognition",
40+
"Object recognition", "3D scans", "Machine learning",
41+
"Filtering", "Deepfake", "Camera", "Cameraphone",
42+
"Interactive panoramas", "Non-Visible Spectrum", "UGV",
43+
"Webcams", "Satellite images")),
44+
TechUsed= col_factor(levels = c(
45+
"Holograms", "Augmented reality", "Ocular implant",
46+
"Emotion recognition", "Surveillance cameras", "AI",
47+
"Virtual reality", "Motion tracking", "Body scans",
48+
"Drones", "MicroscopeOrTelescope", "Biometrics",
49+
"Image generation", "Facial recognition",
50+
"Object recognition", "3D scans", "Machine learning",
51+
"Filtering", "Deepfake", "Camera", "Cameraphone",
52+
"Interactive panoramas", "Non-Visible Spectrum", "UGV",
53+
"Webcams", "Satellite images"))))
54+
55+
56+
# FIGURE 1: Line graph showing genres by year -------------------------------------
57+
58+
CreativeWorks %>%
59+
select(WorkID, Genre, Year) %>%
60+
distinct() %>%
61+
filter(Year >2000) %>%
62+
ggplot(aes(x=Year, fill=Genre, colour = Genre)) +
63+
geom_line(stat="count") +
64+
scale_colour_tableau() +
65+
labs(fill="",
66+
#title ="Year of publication or first release for creative works in the machine vision dataset",
67+
y = "",
68+
x = "") +
69+
theme_minimal()
70+
71+
# FIGURE 2: Plot geographic distribution --------------------------------------------
72+
73+
# See https://cran.r-project.org/web/packages/countrycode/countrycode.pdf
74+
# for description of this package's groupings of countries.
75+
# setting destination = "region" means
76+
# sorting countries into 7 regions as defined by the World Bank
77+
# development Indicators.
78+
#
79+
library(countrycode)
80+
81+
CreativeWorks$Continent <- countrycode(
82+
sourcevar = CreativeWorks$Country,
83+
origin = "country.name",
84+
destination = "region")
85+
86+
CreativeWorks$Continent = factor(CreativeWorks$Continent,
87+
levels = c("North America", "Europe & Central Asia",
88+
"East Asia & Pacific", "Latin America & Caribbean",
89+
"Sub-Saharan Africa","Middle East & North Africa",
90+
"South Asia"))
91+
92+
# Grouping regions with few cases, then plotting.
93+
#
94+
CreativeWorks %>%
95+
select(WorkID, Genre, Continent) %>%
96+
distinct() %>%
97+
select(Genre, Continent) %>%
98+
mutate(Continent = recode(Continent,
99+
"Middle East & North Africa" = "Middle East & Africa",
100+
"Sub-Saharan Africa" = "Middle East & Africa",
101+
"East Asia & Pacific" = "East & South Asia & Pacific",
102+
"South Asia" = "East & South Asia & Pacific")) %>%
103+
ggplot(aes(x=fct_infreq(Continent), fill = Genre)) +
104+
geom_bar() +
105+
geom_text(stat = "count",
106+
aes(label =..count..),
107+
position = position_stack(vjust = 0.5),
108+
size=3,
109+
colour="white") +
110+
scale_fill_tableau() +
111+
labs(
112+
#title ="Geographic distribution of creative works in the machine vision dataset",
113+
#subtitle = "Note that some works are affiliated with several countries.",
114+
y = "Number of works",
115+
x = "") +
116+
theme_minimal() +
117+
theme(axis.text.x = element_text(angle=45, vjust=1, hjust = 1)) +
118+
facet_wrap(~Genre, scales = "free_x")
119+
120+
121+
# Stacked barchart of geographic distribution by continent ----------------
122+
123+
CreativeWorks %>%
124+
select(WorkID, Genre, Continent) %>%
125+
distinct() %>%
126+
select(Genre, Continent) %>%
127+
mutate(Continent = recode(Continent,
128+
"Middle East & North Africa" = "Middle East & Africa",
129+
"Sub-Saharan Africa" = "Middle East & Africa",
130+
"East Asia & Pacific" = "East & South Asia & Pacific",
131+
"South Asia" = "East & South Asia & Pacific")) %>%
132+
pivot_longer(-Continent) %>%
133+
ggplot(aes(x=Continent, fill=value)) +
134+
geom_bar() +
135+
geom_text(stat = "count",
136+
aes(label =..count..),
137+
position = position_stack(vjust = 0.5),
138+
size=4,
139+
colour="white") +
140+
scale_fill_tableau() +
141+
labs(fill="",
142+
title ="Geographic distribution of creative works in the machine vision dataset",
143+
subtitle = "Note that some works are affiliated with several countries.",
144+
y = "",
145+
x = "") +
146+
theme_minimal() +
147+
theme(axis.text.x = element_text(angle=45, vjust=1, hjust = 1))
148+
149+
150+
# Distribution by country in Europe ---------------------------------------
151+
152+
# Set levels for Country so we can sort them by frequency
153+
# using fct_infreq() in ggplot().
154+
levels(CreativeWorks$Country) = factor(unique(CreativeWorks$Country))
155+
156+
# Plot barchart for Europe and Central Asia
157+
CreativeWorks %>%
158+
select(WorkID, Genre, Continent, Country) %>%
159+
distinct() %>%
160+
filter(Continent == "Europe & Central Asia") %>%
161+
select(Genre, Country) %>%
162+
pivot_longer(-Country) %>%
163+
ggplot(aes(x=fct_infreq(Country), fill=value)) +
164+
geom_bar() +
165+
geom_text(stat = "count",
166+
aes(label =..count..),
167+
position = position_stack(vjust = 0.5),
168+
size=3,
169+
colour="white") +
170+
scale_fill_tableau() +
171+
labs(fill="",
172+
title ="European and Central Asian works in the Machine Vision dataset",
173+
subtitle = "Note that some works are affiliated with several countries.",
174+
y = "",
175+
x = "") +
176+
theme_minimal() +
177+
theme(axis.text.x = element_text(angle=45, vjust=1, hjust = 1))
178+
179+
180+
# UK vs everything else ---------------------------------------------------
181+
182+
CreativeWorks$Region <- countrycode(
183+
sourcevar = CreativeWorks$Country,
184+
origin = "country.name",
185+
destination = "un.regionsub.name")
186+
187+
CreativeWorks %>%
188+
select(WorkID, Genre, Continent, Region, Country) %>%
189+
distinct() %>%
190+
filter(Continent == "Europe & Central Asia" & Country != "Kazakhstan") %>%
191+
mutate(Region1 = recode(Country,
192+
"United Kingdom" = "UK & Ireland",
193+
"Ireland" = "UK & Ireland",
194+
"Russia" = "Russia",
195+
.default = Region)) %>%
196+
select(Genre, Region1) %>%
197+
pivot_longer(-Region1) %>%
198+
ggplot(aes(x=fct_infreq(Region1), fill=value)) +
199+
geom_bar() +
200+
geom_text(stat = "count",
201+
aes(label =..count..),
202+
position = position_stack(vjust = 0.5),
203+
size=3,
204+
colour="white") +
205+
scale_fill_tableau() +
206+
labs(fill="",
207+
title ="European and Central Asian works in the Machine Vision dataset",
208+
subtitle = "Note that some works are affiliated with several countries.",
209+
y = "",
210+
x = "") +
211+
theme_minimal() +
212+
theme(axis.text.x = element_text(angle=45, vjust=1, hjust = 1))
213+
214+
215+
view(
216+
CreativeWorks %>%
217+
select(WorkID, Genre, Continent, Region, Country) %>%
218+
distinct() %>%
219+
filter(Continent == "Europe & Central Asia"))
220+
221+
# Distribution by year ----------------------------------------------------
222+
223+
CreativeWorks %>%
224+
select(WorkID, Genre, Year) %>%
225+
distinct() %>%
226+
filter(Year >2000) %>%
227+
ggplot(aes(x=Year, fill=Genre)) +
228+
geom_bar(aes(x = Year, fill = Genre)) +
229+
geom_text(stat = "count",
230+
aes(label =..count..),
231+
position = position_stack(vjust = 0.5),
232+
size=3,
233+
colour="white") +
234+
scale_fill_tableau() +
235+
labs(fill="",
236+
title ="Year of publication or first release for creative works in the machine vision dataset",
237+
y = "",
238+
x = "")
239+
240+
241+

0 commit comments

Comments
 (0)