Set Up
library(tidyverse)
## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr 1.1.4 ✔ readr 2.1.5
## ✔ forcats 1.0.0 ✔ stringr 1.5.1
## ✔ ggplot2 3.5.1 ✔ tibble 3.2.1
## ✔ lubridate 1.9.3 ✔ tidyr 1.3.1
## ✔ purrr 1.0.2
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors
library(broom) # tidy model results
library(umap) # dimension reduction
## Warning: package 'umap' was built under R version 4.4.3
library(plotly) # interactive visualization
##
## Attaching package: 'plotly'
##
## The following object is masked from 'package:ggplot2':
##
## last_plot
##
## The following object is masked from 'package:stats':
##
## filter
##
## The following object is masked from 'package:graphics':
##
## layout
employed <- read_csv("https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2021/2021-02-23/employed.csv")
## Rows: 8184 Columns: 7
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (4): industry, major_occupation, minor_occupation, race_gender
## dbl (3): industry_total, employ_n, year
##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
1 Convert data to standardized form
employed_grouped <- employed %>%
filter(!is.na(employ_n)) %>%
group_by(occupation = paste(industry, minor_occupation), race_gender) %>%
summarise(n = sum(employ_n)) %>%
ungroup()
## `summarise()` has grouped output by 'occupation'. You can override using the
## `.groups` argument.
employed_tidy <- employed_grouped %>%
# Remove total category
filter(race_gender !="Total") %>%
# Add total column
left_join(employed_grouped %>%
filter(race_gender == "TOTAL") %>%
select(occupation, total = n)) %>%
# Get pct in total
mutate(pct = n / total) %>%
# Remove Outliers
filter(total > 1000) %>%
select(-n)
## Joining with `by = join_by(occupation)`
employed_tidy
## # A tibble: 1,392 × 4
## occupation race_gender total pct
## <chr> <chr> <dbl> <dbl>
## 1 Agriculture and related Construction and extractio… Asian 7.3 e4 0.0274
## 2 Agriculture and related Construction and extractio… Black or A… 7.3 e4 0.0822
## 3 Agriculture and related Construction and extractio… Men 7.3 e4 0.973
## 4 Agriculture and related Construction and extractio… TOTAL 7.3 e4 1
## 5 Agriculture and related Construction and extractio… White 7.3 e4 0.863
## 6 Agriculture and related Construction and extractio… Women 7.3 e4 0.0274
## 7 Agriculture and related Farming, fishing, and fore… Asian 5.74e6 0.0139
## 8 Agriculture and related Farming, fishing, and fore… Black or A… 5.74e6 0.0342
## 9 Agriculture and related Farming, fishing, and fore… Men 5.74e6 0.789
## 10 Agriculture and related Farming, fishing, and fore… TOTAL 5.74e6 1
## # ℹ 1,382 more rows
employed_standard <- employed_tidy %>%
# Standardize
group_by(race_gender) %>%
mutate(pct = pct %>% scale() %>% as.numeric()) %>%
ungroup()%>%
mutate(total = total %>% log() %>% scale() %>% as.numeric())
employed_standard
## # A tibble: 1,392 × 4
## occupation race_gender total pct
## <chr> <chr> <dbl> <dbl>
## 1 Agriculture and related Construction and extracti… Asian -1.30 -0.539
## 2 Agriculture and related Construction and extracti… Black or A… -1.30 -0.405
## 3 Agriculture and related Construction and extracti… Men -1.30 1.31
## 4 Agriculture and related Construction and extracti… TOTAL -1.30 NaN
## 5 Agriculture and related Construction and extracti… White -1.30 0.725
## 6 Agriculture and related Construction and extracti… Women -1.30 -1.30
## 7 Agriculture and related Farming, fishing, and for… Asian 0.819 -0.928
## 8 Agriculture and related Farming, fishing, and for… Black or A… 0.819 -1.21
## 9 Agriculture and related Farming, fishing, and for… Men 0.819 0.510
## 10 Agriculture and related Farming, fishing, and for… TOTAL 0.819 NaN
## # ℹ 1,382 more rows
4 Select Optimal Number of CLusters
kclusts <- tibble(k = 1:9) %>%
mutate(kclust = map(.x = k, .f = ~ kmeans(occupation_demo_tbl %>%
select(-occupation, -total_2), centers = .x, nstart = 20)),
glanced = map(.x = kclust, .f = glance))
kclusts %>%
unnest(glanced) %>%
ggplot(aes(k, tot.withinss)) +
geom_point() +
geom_line()
final_cluster <- kmeans(occupation_demo_tbl %>% select(-occupation, -total_2), centers = 5, nstart = 20)
5 Reduce dimension using UMAP
umap_results <- occupation_demo_tbl %>%
select(-occupation, -total_2) %>%
umap()
umap_results_tbl <- umap_results$layout %>%
as.tibble() %>%
bind_cols(occupation_demo_tbl %>% select(occupation))
## Warning: `as.tibble()` was deprecated in tibble 2.0.0.
## ℹ Please use `as_tibble()` instead.
## ℹ The signature and semantics have changed, see `?as_tibble`.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.
## Warning: The `x` argument of `as_tibble.matrix()` must have unique column names if
## `.name_repair` is omitted as of tibble 2.0.0.
## ℹ Using compatibility `.name_repair`.
## ℹ The deprecated feature was likely used in the tibble package.
## Please report the issue at <https://github.com/tidyverse/tibble/issues>.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.
umap_results_tbl
## # A tibble: 232 × 3
## V1 V2 occupation
## <dbl> <dbl> <chr>
## 1 -6.72 -7.34 Agriculture and related Construction and extraction occupatio…
## 2 -0.641 7.01 Agriculture and related Farming, fishing, and forestry occupa…
## 3 -2.28 -5.44 Agriculture and related Installation, maintenance, and repair…
## 4 7.66 -0.915 Agriculture and related Manage-ment, business, and financial …
## 5 0.0481 7.16 Agriculture and related Management, business, and financial o…
## 6 2.52 -4.56 Agriculture and related Office and administrative support occ…
## 7 -2.12 -5.29 Agriculture and related Production occupations
## 8 -0.184 -5.30 Agriculture and related Professional and related occupations
## 9 -6.56 -7.48 Agriculture and related Protective service occupations
## 10 -6.20 -7.73 Agriculture and related Sales and related occupations
## # ℹ 222 more rows
umap_results_tbl %>%
ggplot(aes(V1, V2)) +
geom_point()
6 Visualize Clusters by adding k-means results
kmeans_umap_tbl <- final_cluster %>%
augment(occupation_demo_tbl) %>%
select(occupation, .cluster) %>%
# Add umap results
left_join(umap_results_tbl) %>%
# Add employment info
left_join(employed_tidy %>%
select(-total) %>%
pivot_wider(names_from = race_gender, values_from = pct) %>%
janitor::clean_names())
## Joining with `by = join_by(occupation)`
## Joining with `by = join_by(occupation)`
kmeans_umap_tbl
## # A tibble: 232 × 10
## occupation .cluster V1 V2 asian black_or_african_ame…¹ men total
## <chr> <fct> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 Agricultu… 1 -6.72 -7.34 0.0274 0.0822 0.973 1
## 2 Agricultu… 4 -0.641 7.01 0.0139 0.0342 0.789 1
## 3 Agricultu… 1 -2.28 -5.44 0.0155 0.0309 0.985 1
## 4 Agricultu… 1 7.66 -0.915 0.00992 0.00794 0.739 1
## 5 Agricultu… 4 0.0481 7.16 0.00997 0.00882 0.741 1
## 6 Agricultu… 1 2.52 -4.56 0.0233 0.0155 0.159 1
## 7 Agricultu… 1 -2.12 -5.29 0.0332 0.104 0.815 1
## 8 Agricultu… 1 -0.184 -5.30 0.0339 0.0373 0.675 1
## 9 Agricultu… 1 -6.56 -7.48 0 0.0682 0.864 1
## 10 Agricultu… 1 -6.20 -7.73 0 0.0213 0.585 1
## # ℹ 222 more rows
## # ℹ abbreviated name: ¹black_or_african_american
## # ℹ 2 more variables: white <dbl>, women <dbl>
g <- kmeans_umap_tbl %>%
# Create text label
mutate(text_label = str_glue("Occupation: {occupation}
Cluster: {.cluster}
Asian: {asian %>% scales::percent(1)}
Women: {women %>% scales::percent(1)}")) %>%
# Plot
ggplot(aes(V1, V2, color = .cluster, text = text_label)) +
geom_point()
g %>% ggplotly(tooltip = "text")
