# Import Data
library(tidyverse)
## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr 1.1.3 ✔ readr 2.1.4
## ✔ forcats 1.0.0 ✔ stringr 1.5.0
## ✔ ggplot2 3.4.3 ✔ tibble 3.2.1
## ✔ lubridate 1.9.3 ✔ tidyr 1.3.0
## ✔ 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
data <- readr::read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2020/2020-09-22/members.csv')
## Rows: 76519 Columns: 21
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (10): expedition_id, member_id, peak_id, peak_name, season, sex, citizen...
## dbl (5): year, age, highpoint_metres, death_height_metres, injury_height_me...
## lgl (6): hired, success, solo, oxygen_used, died, injured
##
## 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.
data %>% skimr::skim()
| Name | Piped data |
| Number of rows | 76519 |
| Number of columns | 21 |
| _______________________ | |
| Column type frequency: | |
| character | 10 |
| logical | 6 |
| numeric | 5 |
| ________________________ | |
| Group variables | None |
Variable type: character
| skim_variable | n_missing | complete_rate | min | max | empty | n_unique | whitespace |
|---|---|---|---|---|---|---|---|
| expedition_id | 0 | 1.00 | 9 | 9 | 0 | 10350 | 0 |
| member_id | 0 | 1.00 | 12 | 12 | 0 | 76518 | 0 |
| peak_id | 0 | 1.00 | 4 | 4 | 0 | 391 | 0 |
| peak_name | 15 | 1.00 | 4 | 25 | 0 | 390 | 0 |
| season | 0 | 1.00 | 6 | 7 | 0 | 5 | 0 |
| sex | 2 | 1.00 | 1 | 1 | 0 | 2 | 0 |
| citizenship | 10 | 1.00 | 2 | 23 | 0 | 212 | 0 |
| expedition_role | 21 | 1.00 | 4 | 25 | 0 | 524 | 0 |
| death_cause | 75413 | 0.01 | 3 | 27 | 0 | 12 | 0 |
| injury_type | 74807 | 0.02 | 3 | 27 | 0 | 11 | 0 |
Variable type: logical
| skim_variable | n_missing | complete_rate | mean | count |
|---|---|---|---|---|
| hired | 0 | 1 | 0.21 | FAL: 60788, TRU: 15731 |
| success | 0 | 1 | 0.38 | FAL: 47320, TRU: 29199 |
| solo | 0 | 1 | 0.00 | FAL: 76398, TRU: 121 |
| oxygen_used | 0 | 1 | 0.24 | FAL: 58286, TRU: 18233 |
| died | 0 | 1 | 0.01 | FAL: 75413, TRU: 1106 |
| injured | 0 | 1 | 0.02 | FAL: 74806, TRU: 1713 |
Variable type: numeric
| skim_variable | n_missing | complete_rate | mean | sd | p0 | p25 | p50 | p75 | p100 | hist |
|---|---|---|---|---|---|---|---|---|---|---|
| year | 0 | 1.00 | 2000.36 | 14.78 | 1905 | 1991 | 2004 | 2012 | 2019 | ▁▁▁▃▇ |
| age | 3497 | 0.95 | 37.33 | 10.40 | 7 | 29 | 36 | 44 | 85 | ▁▇▅▁▁ |
| highpoint_metres | 21833 | 0.71 | 7470.68 | 1040.06 | 3800 | 6700 | 7400 | 8400 | 8850 | ▁▁▆▃▇ |
| death_height_metres | 75451 | 0.01 | 6592.85 | 1308.19 | 400 | 5800 | 6600 | 7550 | 8830 | ▁▁▂▇▆ |
| injury_height_metres | 75510 | 0.01 | 7049.91 | 1214.24 | 400 | 6200 | 7100 | 8000 | 8880 | ▁▁▂▇▇ |
Goal: Predict the death on Nepal Himalaya climbers
Issues with data:
Missing values
Factors or numeric variables:
Zero variance variables
Character variables: Convert to numbers in the recipies step: expedition_id, member_id, peak_id, peak_name, citizenship, expedition_role, death_cause, injury_type
Unbalanced target variable: died
ID variable: member_id
# Treating missing values
data_clean <- data %>%
select(-death_cause, -injury_type, -death_height_metres, -injury_height_metres) %>%
na.omit()
data_clean %>% count(died)
## # A tibble: 2 × 2
## died n
## <lgl> <int>
## 1 FALSE 51639
## 2 TRUE 744
data_clean %>%
ggplot(aes(died)) +
geom_bar()
Died vs. expedition_role
top_10_exp_role_vec <- data_clean %>%
count(expedition_role, sort = TRUE) %>%
head(10) %>%
pull(expedition_role)
# Relationship between pay and attrition
data_clean %>%
filter(expedition_role %in% top_10_exp_role_vec) %>%
count(died, expedition_role) %>%
ggplot(aes(died, expedition_role, fill = n)) +
geom_tile()
Relationship in all variables with Correlation plot
library(correlationfunnel)
## ══ correlationfunnel Tip #3 ════════════════════════════════════════════════════
## Using `binarize()` with data containing many columns or many rows can increase dimensionality substantially.
## Try subsetting your data column-wise or row-wise to avoid creating too many columns.
## You can always make a big problem smaller by sampling. :)
# Step 1: Binarize
data_binarized <- data_clean %>%
select(-member_id) %>% # ID variable
binarize()
data_binarized %>% glimpse()
## Rows: 52,383
## Columns: 82
## $ expedition_id__HIML13308 <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ `expedition_id__-OTHER` <dbl> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, …
## $ peak_id__AMAD <dbl> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, …
## $ peak_id__ANN1 <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ peak_id__BARU <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ peak_id__CHOY <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ peak_id__DHA1 <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ peak_id__EVER <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ peak_id__HIML <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ peak_id__KANG <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ peak_id__LHOT <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ peak_id__MAKA <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ peak_id__MANA <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ peak_id__PUMO <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ `peak_id__-OTHER` <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ peak_name__Ama_Dablam <dbl> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, …
## $ peak_name__Annapurna_I <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ peak_name__Baruntse <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ peak_name__Cho_Oyu <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ peak_name__Dhaulagiri_I <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ peak_name__Everest <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ peak_name__Himlung_Himal <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ peak_name__Kangchenjunga <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ peak_name__Lhotse <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ peak_name__Makalu <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ peak_name__Manaslu <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ peak_name__Pumori <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ `peak_name__-OTHER` <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ `year__-Inf_1997` <dbl> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, …
## $ year__1997_2007 <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ year__2007_2012 <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ year__2012_Inf <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ season__Autumn <dbl> 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ season__Spring <dbl> 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, …
## $ season__Winter <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ `season__-OTHER` <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ sex__F <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ sex__M <dbl> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, …
## $ `age__-Inf_29` <dbl> 0, 0, 1, 0, 1, 0, 1, 0, 0, 1, 0, 0, 1, …
## $ age__29_36 <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 1, 0, …
## $ age__36_43 <dbl> 1, 1, 0, 1, 0, 1, 0, 1, 0, 0, 0, 0, 0, …
## $ age__43_Inf <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ citizenship__Australia <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ citizenship__Austria <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ citizenship__Canada <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ citizenship__China <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ citizenship__France <dbl> 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ citizenship__Germany <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ citizenship__India <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ citizenship__Italy <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ citizenship__Japan <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ citizenship__Nepal <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ citizenship__New_Zealand <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ citizenship__Poland <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ citizenship__Russia <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ citizenship__S_Korea <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ citizenship__Spain <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ citizenship__Switzerland <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ citizenship__UK <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ citizenship__USA <dbl> 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, …
## $ `citizenship__-OTHER` <dbl> 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, …
## $ expedition_role__Climber <dbl> 0, 0, 1, 1, 1, 1, 1, 0, 0, 1, 0, 1, 0, …
## $ expedition_role__Deputy_Leader <dbl> 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ `expedition_role__H-A_Worker` <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ expedition_role__Leader <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ `expedition_role__-OTHER` <dbl> 0, 1, 0, 0, 0, 0, 0, 1, 1, 0, 1, 0, 1, …
## $ hired__0 <dbl> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, …
## $ hired__1 <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ `highpoint_metres__-Inf_6750` <dbl> 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ highpoint_metres__6750_7400 <dbl> 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, …
## $ highpoint_metres__7400_8450 <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ highpoint_metres__8450_Inf <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ success__0 <dbl> 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ success__1 <dbl> 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, …
## $ solo__0 <dbl> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, …
## $ `solo__-OTHER` <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ oxygen_used__0 <dbl> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, …
## $ oxygen_used__1 <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ died__0 <dbl> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, …
## $ died__1 <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ injured__0 <dbl> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, …
## $ injured__1 <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
# Step 2: Correlation
data_correlation <- data_binarized %>%
correlate(died__1)
## Warning: correlate(): [Data Imbalance Detected] Consider sampling to balance the classes more than 5%
## Column with imbalance: died__1
data_correlation
## # A tibble: 82 × 3
## feature bin correlation
## <fct> <chr> <dbl>
## 1 died 0 -1
## 2 died 1 1
## 3 year -Inf_1997 0.0843
## 4 success 0 0.0562
## 5 success 1 -0.0562
## 6 peak_id ANN1 0.0431
## 7 peak_name Annapurna_I 0.0431
## 8 year 2012_Inf -0.0330
## 9 peak_id AMAD -0.0323
## 10 peak_name Ama_Dablam -0.0323
## # ℹ 72 more rows
# Step 3: Plot
data_correlation %>%
correlationfunnel::plot_correlation_funnel()
## Warning: ggrepel: 45 unlabeled data points (too many overlaps). Consider
## increasing max.overlaps