RStudio Guide

To get started, click on the 'workspace' folder in the bottom right pane of RStudio. Click on the file entitled "Step 1...". You may want to drag the console pane to be smaller so that you have more room to work. You'll complete each task for Step 1 in that R Markdown file. Remember, you must run the cells with the play button at the top right of each cell for a task before moving onto the next task in the R Markdown file. Continue until you have completed all tasks in this step. Then when you are ready to move onto the next step, you'll come back and click on the file for the next step until you have completed all tasks in all steps of the lab.

Exploring High-Dimensionality Data

To review the concepts covered in this step, please refer to the Understanding the Importance of Reducing Complexity in Data module of the Performing Dimension Analysis with R course.

Understanding the importance of reducing complexity in data is crucial because it sets the foundation for effective dimensionality reduction and feature selection. This step will help learners grasp the problems associated with high-dimensionality data and the difference between feature selection and dimensionality reduction.

Dive into the world of data by exploring the HRIS.csv dataset. Though this dataset is relatively small, the techniques you learn will apply to datasets of any size. Using R, you'll perform exploratory data analysis (EDA) to identify key aspects of the data. Tools such as ggplot2 for visualization and dplyr for data manipulation will be essential in this step. Objective: Gain a comprehensive understanding of high-dimensional data characteristics and the necessity for dimensionality reduction.

Task 1.1: Load Required Libraries

Before diving into the data, you need to load the necessary libraries for data manipulation and visualization. For this task, load the dplyr and ggplot2 libraries.

library(dplyr)
library(ggplot2)

Task 1.2: Read and Inspect the Dataset

Now that the libraries are loaded, the next step is to read the HRIS.csv file into R as hr_data and inspect the first few rows of the dataset to understand its structure.

hr_data <- read.csv('HRIS.csv')
head(hr_data)

Task 1.3: Explore Data Dimensions

Understanding the dimensions of your dataset is crucial for recognizing high-dimensionality. Use R to find out the number of rows and columns in the hr_data dataset.

dim(hr_data)

Task 1.4: Summarize the Dataset

Get a quick summary of the dataset to understand the distribution of data across different columns. This will help in identifying any potential issues with the data.

summary(hr_data)

Task 1.5: Visualize Salary Distribution

Visualizing data can provide insights that are not immediately obvious from the raw data. Create a histogram to visualize the distribution of salaries in the dataset.

ggplot(hr_data, aes(x = salary)) +
  geom_histogram(binwidth = 5000)

Implementing Principal Components Analysis (PCA)

To review the concepts covered in this step, please refer to the Performing Dimensional Analysis for Continuous Data module of the Performing Dimension Analysis with R course.

Principal Components Analysis (PCA) is important because it's a fundamental technique for reducing dimensionality in linear data, helping to mitigate the curse of dimensionality. This step will provide hands-on experience in applying PCA to continuous data.

In this step, you'll apply Principal Components Analysis (PCA) to the HRIS.csv dataset to reduce its dimensionality. The goal is to extract the most significant components that explain the majority of the variance in the data. You'll use the prcomp function in R. Objective: Learn how to perform PCA on a dataset and interpret the results to understand the underlying structure of the data.

Task 2.1: Loading and Inspecting the HRIS Dataset

Load the HRIS.csv dataset into R as hr_data and inspect the first few rows to understand its structure.

# Load the HRIS dataset
hr_data <- read.csv('HRIS.csv')

# Display the first few rows of the dataset
head(hr_data)

Task 2.2: Preprocessing Data for PCA

Before performing PCA, it's important to preprocess the data. Load the dplyr package to help with preprocessing.

For PCA, only meaningful numeric variables are appropriate. Subset the data to columns that fit this description.

# Load the dplyr package
library(dplyr)

# Retain only meaningful numeric variables
hr_data_numeric <- hr_data %>%
  select(salary, benefits, latitude, longitude, employee_level, tenure_in_days)

Task 2.3: Performing PCA

Perform a principal componenets analysis (PCA) on the dataset. Specify that the variables should be scaled to unit variance (i.e., standardized). Summarize the result. How much cumulative variance is explained by 3 principal components?

# Perform PCA
pca <- prcomp(hr_data_numeric, scale. = TRUE)
summary(pca)

# ___% of variance is explained by 3 components
# ~82%

Task 2.4: Visualize with a Biplot

Use the biplot function to plot the PCA results. Which variables are most related to the first principle component (the x-axis)?

# Make a biplot
biplot(pca)

# The variables ____ are most related to the first principal component
# salary, employee_level, benefits

Applying Linear Discriminant Analysis (LDA)

To review the concepts covered in this step, please refer to the Performing Dimensional Analysis for Categorical Data module of the Performing Dimension Analysis with R course.

Linear Discriminant Analysis (LDA) is important because it provides a method for dimensionality reduction that is particularly useful for categorical data. This step will offer practical experience in using LDA for dimensionality reduction.

This step involves applying Linear Discriminant Analysis (LDA) to the HRIS.csv dataset to reduce its dimensionality while considering the categorical nature of some of its variables. You'll convert appropriate columns to factors and use the MASS package's lda function to perform LDA. Objective: Understand how to apply LDA to a dataset with categorical variables and interpret the results.

Task 3.1: Loading and Inspecting the Dataset

Begin by loading the HRIS.csv dataset into R as hr_data. Use appropriate functions to inspect the dataset, focusing on its structure and the first few rows to understand the data you'll be working with.

# Load the HRIS.csv dataset
hr_data <- read.csv('HRIS.csv')

# Inspect the structure of the dataset
str(hr_data)

# Display the first few rows of the dataset
head(hr_data)

Task 3.2: Converting Categorical Variables to Factors

Some columns in the dataset are categorical and should be treated as factors for the analysis. Convert the gender, position, and department columns to factors.

# Convert the specified columns to factors
hr_data$gender <- factor(hr_data$gender)
hr_data$position <- factor(hr_data$position)
hr_data$department <- factor(hr_data$department)

Task 3.3: Performing LDA

Load the MASS package and perform linear discriminant analysis (LDA). Predict an employee's position using their salary, benefits, and employee_level. Print the results.

library(MASS)

lda_model <- lda(formula = position ~ salary + benefits + employee_level, data = hr_data)
lda_model

Exploring Manifold Learning Techniques

To review the concepts covered in this step, please refer to the Performing Dimensional Analysis for Non-linear Data module of the Performing Dimension Analysis with R course.

Manifold Learning is crucial for dealing with non-linear data, as it can uncover the underlying structure of data that is not apparent in high-dimensional space. This step will introduce learners to t-SNE, a manifold learning technique.

Embark on a journey through non-linear data by applying t-SNE to the HRIS.csv dataset. The goal is to discover low-dimensional manifolds that represent the data's structure. You'll use the Rtsne and dplyr packages. Objective: Gain practical experience in applying manifold learning techniques to non-linear data and understand how to interpret the results.

Task 4.1: Loading the HRIS Dataset

Begin by loading the HRIS.csv dataset into R as hr_data. This dataset contains various employee details, which you will use to explore manifold learning techniques. Use the read.csv function to load the dataset into a variable named hr_data.

hr_data <- read.csv('HRIS.csv')

Task 4.2: Applying t-SNE

Load the dplyr and Rtsne R packages, which are already installed in this environment. Subset the data to meaningful numeric columns. Then apply the t-SNE technique to reduce the dimensionality of the data to 2. Then, apply t-SNE to the dataset and store the result in a variable named tsne_results. Finally, plot the results.

# Load the dplyr and Rtsne packages
library(dplyr)
library(Rtsne)

# Retain only meaningful numeric variables
hr_data_numeric <- hr_data %>%
  dplyr::select(salary, benefits, latitude, longitude, employee_level, tenure_in_days)

# Apply t-SNE to the HRIS dataset
tsne_results <- Rtsne(hr_data_numeric, dims = 2)

# Plot the t-SNE results
plot(tsne_results$Y)