VLU Workshop on Machine Learning - Day 1
Thach Tran
2023-06-07
Van Lang University Workshop on Machine Learning - Day 1: Data visualization
Việc 1. Tải R, RStudio về máy tính
https://cran.r-project.org/mirrors.html https://www.rstudio.com/products/rstudio/download
Việc 2. Cài đặt các packages cần thiết
#install.packages(c("readxl","tidyverse","ggplot2","GGally","DescTools","table1","compareGroups","caret","MASS","pROC","factoextra","devtools","datasets","ape","ggdendro","NbClust","dendextend","cluster","randomForest"), dependencies = T)Việc 3. Đọc dữ liệu “Professorial Salaries” vào R
library(readxl)
salary = read_excel("C:\\Thach\\VLU workshop (Jun2023)\\Datasets\\Professorial Salaries.xlsx")việc 4. Thông tin về dữ liệu salary
dim(salary)## [1] 397 7head(salary)## # A tibble: 6 × 7
## ID Rank Discipline Yrs.since.phd Yrs.service Sex Salary
## <dbl> <chr> <chr> <dbl> <dbl> <chr> <dbl>
## 1 1 Prof B 19 18 Male 139750
## 2 2 Prof B 20 16 Male 173200
## 3 3 AsstProf B 4 3 Male 79750
## 4 4 Prof B 45 39 Male 115000
## 5 5 Prof B 40 41 Male 141500
## 6 6 AssocProf B 6 6 Male 97000tail(salary)## # A tibble: 6 × 7
## ID Rank Discipline Yrs.since.phd Yrs.service Sex Salary
## <dbl> <chr> <chr> <dbl> <dbl> <chr> <dbl>
## 1 392 Prof A 30 19 Male 151292
## 2 393 Prof A 33 30 Male 103106
## 3 394 Prof A 31 19 Male 150564
## 4 395 Prof A 42 25 Male 101738
## 5 396 Prof A 25 15 Male 95329
## 6 397 AsstProf A 8 4 Male 81035Việc 5. Biên tập dữ liệu với package tidyverse
library(tidyverse)## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr 1.1.0 ✔ readr 2.1.4
## ✔ forcats 1.0.0 ✔ stringr 1.5.0
## ✔ ggplot2 3.4.1 ✔ tibble 3.2.1
## ✔ lubridate 1.9.2 ✔ tidyr 1.3.0
## ✔ purrr 1.0.1
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()
## ℹ Use the ]8;;http://conflicted.r-lib.org/conflicted package]8;; to force all conflicts to become errorssalary = salary %>% mutate(salary.level = case_when(Salary >= 150000~ "High",
Salary>100000 & Salary< 150000~ "Medium",
Salary<=100000~ "Low"))
salary = salary %>% mutate(salary.vn = Salary*23481532)
head(salary)## # A tibble: 6 × 9
## ID Rank Discipline Yrs.since.phd Yrs.service Sex Salary salary.level
## <dbl> <chr> <chr> <dbl> <dbl> <chr> <dbl> <chr>
## 1 1 Prof B 19 18 Male 139750 Medium
## 2 2 Prof B 20 16 Male 173200 High
## 3 3 AsstProf B 4 3 Male 79750 Low
## 4 4 Prof B 45 39 Male 115000 Medium
## 5 5 Prof B 40 41 Male 141500 Medium
## 6 6 AssocProf B 6 6 Male 97000 Low
## # ℹ 1 more variable: salary.vn <dbl>Men.A.high = salary %>% filter(Sex == "Male", Discipline == "A", salary.level == "High")
head(Men.A.high)## # A tibble: 6 × 9
## ID Rank Discipline Yrs.since.phd Yrs.service Sex Salary salary.level
## <dbl> <chr> <chr> <dbl> <dbl> <chr> <dbl> <chr>
## 1 127 Prof A 28 26 Male 155500 High
## 2 135 Prof A 35 25 Male 168635 High
## 3 140 Prof A 21 18 Male 152664 High
## 4 250 Prof A 29 7 Male 204000 High
## 5 272 Prof A 42 18 Male 194800 High
## 6 278 Prof A 31 27 Male 163200 High
## # ℹ 1 more variable: salary.vn <dbl>Subset = salary %>% select(ID, Rank, Salary, Sex)
head(Subset)## # A tibble: 6 × 4
## ID Rank Salary Sex
## <dbl> <chr> <dbl> <chr>
## 1 1 Prof 139750 Male
## 2 2 Prof 173200 Male
## 3 3 AsstProf 79750 Male
## 4 4 Prof 115000 Male
## 5 5 Prof 141500 Male
## 6 6 AssocProf 97000 MaleViệc 6. Vẽ biểu đồ histogram tiền lương
library(ggplot2)
p = ggplot(data = salary, aes(x = Salary))
p + geom_histogram(fill = "blue", col = "white") + labs(x = "Salary (USD)", y = "Number of people", title = "Distribution of salary in USD")## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
Việc 7. Vẽ biểu đồ hộp so sánh tiền lương giữa nam và nữ
p = ggplot(data = salary, aes(x = Sex, y = Salary, col = Sex))
p + geom_boxplot() + geom_jitter(alpha = 0.2) +
labs(x = "Sex", y = "Salary (USD)", title = "Distribution of salary in USD by sexes")
Việc 8. Vẽ biểu đồ tán xạ đánh giá mối liên quan giữa tiền lương và thời gian sau tiến sĩ
8.1 Chung cho cả nam và nữ
ggplot(data = salary, aes(x = Yrs.since.phd, y = Salary)) + geom_point() + geom_smooth(method="loess") + labs(x = "Time since PhD (years)", y= "Salary (dollars)")## `geom_smooth()` using formula = 'y ~ x'
8.2 Riêng cho nam và nữ trên cùng 1 biểu đồ
(Not sure whether it works)
ggplot(data = salary, aes(x = Yrs.since.phd, y = Salary, col = Sex)) + geom_point() + geom_smooth(method="loess") + labs(x = "Time since PhD (years)", y= "Salary (dollars)")## `geom_smooth()` using formula = 'y ~ x'