apply 6
Cam Paquette
2025-03-16
R Markdown
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 errorslibrary(correlationfunnel)## Warning: package 'correlationfunnel' was built under R version 4.4.2## ══ Using correlationfunnel? ════════════════════════════════════════════════════
## You might also be interested in applied data science training for business.
## </> Learn more at - www.business-science.io </>data <- readr::read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/main/data/2023/2023-08-15/spam.csv')## Rows: 4601 Columns: 7
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (1): yesno
## dbl (6): crl.tot, dollar, bang, money, n000, make
##
## 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.Explore data
skimr::skim(data)| Name | data |
| Number of rows | 4601 |
| Number of columns | 7 |
| _______________________ | |
| Column type frequency: | |
| character | 1 |
| numeric | 6 |
| ________________________ | |
| Group variables | None |
Variable type: character
| skim_variable | n_missing | complete_rate | min | max | empty | n_unique | whitespace |
|---|---|---|---|---|---|---|---|
| yesno | 0 | 1 | 1 | 1 | 0 | 2 | 0 |
Variable type: numeric
| skim_variable | n_missing | complete_rate | mean | sd | p0 | p25 | p50 | p75 | p100 | hist |
|---|---|---|---|---|---|---|---|---|---|---|
| crl.tot | 0 | 1 | 283.29 | 606.35 | 1 | 35 | 95 | 266.00 | 15841.00 | ▇▁▁▁▁ |
| dollar | 0 | 1 | 0.08 | 0.25 | 0 | 0 | 0 | 0.05 | 6.00 | ▇▁▁▁▁ |
| bang | 0 | 1 | 0.27 | 0.82 | 0 | 0 | 0 | 0.32 | 32.48 | ▇▁▁▁▁ |
| money | 0 | 1 | 0.09 | 0.44 | 0 | 0 | 0 | 0.00 | 12.50 | ▇▁▁▁▁ |
| n000 | 0 | 1 | 0.10 | 0.35 | 0 | 0 | 0 | 0.00 | 5.45 | ▇▁▁▁▁ |
| make | 0 | 1 | 0.10 | 0.31 | 0 | 0 | 0 | 0.00 | 4.54 | ▇▁▁▁▁ |
data_clean <- data %>%
mutate(yesno = factor(yesno, levels = c("y", "n")))Comparing the occurrence of the dollar/money sign and how it relates to emails being spam
data_clean %>%
filter(dollar > 0, money > 0) %>%
ggplot(aes(money, dollar, color = yesno)) +
geom_point(alpha = 0.3, size = 1.0)
data_clean %>%
ggplot(aes(dollar, y = after_stat(density), fill = yesno)) +
geom_histogram(position = "identity", alpha = 1.0)## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
data_clean %>%
ggplot(aes(money, y = after_stat(density), fill = yesno)) +
geom_histogram(position = "identity", alpha = 1.0)## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
# Step 1: Binarize
data_binarized <- data_clean %>%
select(-dollar) %>%
binarize()
data_binarized %>% glimpse()## Rows: 4,601
## Columns: 15
## $ `crl.tot__-Inf_35` <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0, 0, 0, …
## $ crl.tot__35_95 <dbl> 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ crl.tot__95_266 <dbl> 0, 0, 0, 1, 1, 0, 1, 0, 0, 0, 0, 1, 1, 0, 1, 1, 1, …
## $ crl.tot__266_Inf <dbl> 1, 1, 1, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, …
## $ `bang__-Inf_0.315` <dbl> 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 0, 0, 0, …
## $ bang__0.315_Inf <dbl> 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0, 1, 1, 1, …
## $ money__0 <dbl> 1, 0, 0, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1, 0, 1, …
## $ `money__-OTHER` <dbl> 0, 1, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 0, …
## $ n000__0 <dbl> 1, 0, 0, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 0, 1, 1, …
## $ `n000__-OTHER` <dbl> 0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, …
## $ make__0 <dbl> 1, 0, 0, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 1, 1, 1, 1, …
## $ make__0.1 <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ `make__-OTHER` <dbl> 0, 1, 1, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, …
## $ yesno__y <dbl> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, …
## $ yesno__n <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …# Step 2: Correlation
data_correlation <- data_binarized %>%
correlate(yesno__y)
data_correlation## # A tibble: 15 × 3
## feature bin correlation
## <fct> <chr> <dbl>
## 1 yesno y 1
## 2 yesno n -1
## 3 bang -Inf_0.315 -0.490
## 4 bang 0.315_Inf 0.490
## 5 money -OTHER 0.475
## 6 money 0 -0.475
## 7 n000 0 -0.419
## 8 n000 -OTHER 0.419
## 9 crl.tot -Inf_35 -0.360
## 10 crl.tot 266_Inf 0.299
## 11 make 0 -0.239
## 12 make -OTHER 0.223
## 13 crl.tot 95_266 0.145
## 14 crl.tot 35_95 -0.0818
## 15 make 0.1 0.0803# Step 3: Plot
data_correlation %>%
correlationfunnel::plot_correlation_funnel()
library(tidymodels)## Warning: package 'tidymodels' was built under R version 4.4.2## ── Attaching packages ────────────────────────────────────── tidymodels 1.2.0 ──## ✔ broom 1.0.6 ✔ rsample 1.2.1
## ✔ dials 1.3.0 ✔ tune 1.2.1
## ✔ infer 1.0.7 ✔ workflows 1.1.4
## ✔ modeldata 1.4.0 ✔ workflowsets 1.1.0
## ✔ parsnip 1.2.1 ✔ yardstick 1.3.2
## ✔ recipes 1.1.0## Warning: package 'dials' was built under R version 4.4.2## Warning: package 'infer' was built under R version 4.4.2## Warning: package 'modeldata' was built under R version 4.4.2## Warning: package 'parsnip' was built under R version 4.4.2## Warning: package 'tune' was built under R version 4.4.2## Warning: package 'workflows' was built under R version 4.4.2## Warning: package 'workflowsets' was built under R version 4.4.3## Warning: package 'yardstick' was built under R version 4.4.2## ── Conflicts ───────────────────────────────────────── tidymodels_conflicts() ──
## ✖ scales::discard() masks purrr::discard()
## ✖ dplyr::filter() masks stats::filter()
## ✖ recipes::fixed() masks stringr::fixed()
## ✖ dplyr::lag() masks stats::lag()
## ✖ yardstick::spec() masks readr::spec()
## ✖ recipes::step() masks stats::step()
## • Use suppressPackageStartupMessages() to eliminate package startup messagesSplit Data
set.seed(1234)
# data_clean <- data_clean %>% sample_n(100)
data_split <- initial_split(data_clean, strata = yesno)
data_train <- training(data_split)
data_test <- testing(data_split)
data_cv <- rsample::vfold_cv(data_train, strata = yesno)
data_cv## # 10-fold cross-validation using stratification
## # A tibble: 10 × 2
## splits id
## <list> <chr>
## 1 <split [3104/346]> Fold01
## 2 <split [3105/345]> Fold02
## 3 <split [3105/345]> Fold03
## 4 <split [3105/345]> Fold04
## 5 <split [3105/345]> Fold05
## 6 <split [3105/345]> Fold06
## 7 <split [3105/345]> Fold07
## 8 <split [3105/345]> Fold08
## 9 <split [3105/345]> Fold09
## 10 <split [3106/344]> Fold10#Preprocess Data
library(themis)## Warning: package 'themis' was built under R version 4.4.3data_rec <- recipes::recipe(yesno ~., data = data_train) %>%
step_YeoJohnson(all_nominal_predictors())
data_rec %>% prep() %>% juice() %>% glimpse()## Rows: 3,450
## Columns: 7
## $ crl.tot <dbl> 26, 65, 1866, 35, 150, 28, 16, 791, 26, 27, 50, 16, 28, 13, 11…
## $ dollar <dbl> 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000,…
## $ bang <dbl> 0.149, 0.262, 0.000, 0.000, 0.000, 0.393, 0.729, 0.149, 0.000,…
## $ money <dbl> 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.…
## $ n000 <dbl> 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.…
## $ make <dbl> 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.…
## $ yesno <fct> n, n, n, n, n, n, n, n, n, n, n, n, n, n, n, n, n, n, n, n, n,…Build Model
library(usemodels)## Warning: package 'usemodels' was built under R version 4.4.2data <- data%>%
mutate(yesno = as.numeric(yesno))## Warning: There was 1 warning in `mutate()`.
## ℹ In argument: `yesno = as.numeric(yesno)`.
## Caused by warning:
## ! NAs introduced by coercionusemodels::use_xgboost(yesno ~ ., data = data_train)## xgboost_recipe <-
## recipe(formula = yesno ~ ., data = data_train) %>%
## step_zv(all_predictors())
##
## xgboost_spec <-
## boost_tree(trees = tune(), min_n = tune(), tree_depth = tune(), learn_rate = tune(),
## loss_reduction = tune(), sample_size = tune()) %>%
## set_mode("classification") %>%
## set_engine("xgboost")
##
## xgboost_workflow <-
## workflow() %>%
## add_recipe(xgboost_recipe) %>%
## add_model(xgboost_spec)
##
## set.seed(96026)
## xgboost_tune <-
## tune_grid(xgboost_workflow, resamples = stop("add your rsample object"), grid = stop("add number of candidate points"))xgboost_spec <-
boost_tree(trees = tune()) %>%
set_mode("classification") %>%
set_engine("xgboost")
xgboost_workflow <-
workflow() %>%
add_recipe(data_rec) %>%
add_model(xgboost_spec) Tune Hyperparameters
doParallel::registerDoParallel()
set.seed(50926)
xgboost_tune <-
tune_grid(xgboost_workflow,
resamples = data_cv,
grid = 5,
control = control_grid(save_pred = TRUE))Model Evaluation
Identify optimal value
collect_metrics(xgboost_tune) ## # A tibble: 15 × 7
## trees .metric .estimator mean n std_err .config
## <int> <chr> <chr> <dbl> <int> <dbl> <chr>
## 1 253 accuracy binary 0.872 10 0.00421 Preprocessor1_Model1
## 2 253 brier_class binary 0.103 10 0.00333 Preprocessor1_Model1
## 3 253 roc_auc binary 0.912 10 0.00344 Preprocessor1_Model1
## 4 601 accuracy binary 0.868 10 0.00368 Preprocessor1_Model2
## 5 601 brier_class binary 0.110 10 0.00314 Preprocessor1_Model2
## 6 601 roc_auc binary 0.904 10 0.00313 Preprocessor1_Model2
## 7 1190 accuracy binary 0.864 10 0.00396 Preprocessor1_Model3
## 8 1190 brier_class binary 0.115 10 0.00312 Preprocessor1_Model3
## 9 1190 roc_auc binary 0.900 10 0.00318 Preprocessor1_Model3
## 10 1512 accuracy binary 0.863 10 0.00419 Preprocessor1_Model4
## 11 1512 brier_class binary 0.117 10 0.00320 Preprocessor1_Model4
## 12 1512 roc_auc binary 0.898 10 0.00327 Preprocessor1_Model4
## 13 1976 accuracy binary 0.862 10 0.00440 Preprocessor1_Model5
## 14 1976 brier_class binary 0.119 10 0.00326 Preprocessor1_Model5
## 15 1976 roc_auc binary 0.897 10 0.00339 Preprocessor1_Model5collect_predictions(xgboost_tune) %>%
group_by(id) %>%
roc_curve(yesno, .pred_y) %>% # Ensure you're using the probability for class 1
autoplot()
xgboost_last <- xgboost_workflow %>%
finalize_workflow(select_best(xgboost_tune, metric = "accuracy")) %>%
last_fit(data_split)## Warning: package 'xgboost' was built under R version 4.4.2collect_metrics(xgboost_last)## # A tibble: 3 × 4
## .metric .estimator .estimate .config
## <chr> <chr> <dbl> <chr>
## 1 accuracy binary 0.871 Preprocessor1_Model1
## 2 roc_auc binary 0.914 Preprocessor1_Model1
## 3 brier_class binary 0.0993 Preprocessor1_Model1collect_predictions(xgboost_last) %>%
yardstick::conf_mat(yesno, .pred_class) %>%
autoplot()
library(vip)## Warning: package 'vip' was built under R version 4.4.3##
## Attaching package: 'vip'## The following object is masked from 'package:utils':
##
## vixgboost_last %>%
workflows::extract_fit_engine() %>%
vip()
Model
Step Yeojohnson accuracy .871 roc_auc .914
Step PCA data set lacks numeric variables
Step nomrlilization/log deleted from code no change