Data Preprocessing in R: Ultimate Tutorial [2021]

In our following data preprocessing in R tutorial, you’ll learn the fundamentals of how to perform data preprocessing. This tutorial requires you to be familiar with the basics of R and programming:

1. Step: Finding and Fixing Issues

We’ll start our data preprocessing in R tutorial by importing the data set first. After all, you can’t preprocess the data if you don’t have the data in the first place.

In our case, the data is stored in the data.csv file in the working directory. You can use the command setwd(“desired location”) and set the working directory.  

Here’s how you’ll start the process:

dataset <- read.csv(“Data.csv”)

Here’s our dataset:

## Country Age Salary Purchased
## 1 France 44 72000 No
## 2 Spain 27 48000 Yes
## 3 Germany 30 54000 No
## 4 Spain 38 61000 No
## 5 Germany 40 NA Yes
## 6 France 35 58000 Yes
## 7 Spain NA 52000 No
## 8 France 48 79000 Yes
## 9 Germany 50 83000 No
## 10 France 37 67000 Yes

As you can see, there are missing values in the Salary and Age columns of our dataset. We have identified the issue present in our dataset so we can now start fixing the same. 

No other issues seem to be present in our dataset so we only have to handle the missing values. We can fix this problem by replacing the NA values with the average values of the respective columns. Here’s how:

dataset$Age <- ifelse(is.na(dataset$Age), 

                      ave(dataset$Age, FUN = function(x) 

                        mean(x, na.rm = TRUE)), 

                      dataset$Age)

dataset$Salary <- ifelse(is.na(dataset$Salary), 

                      ave(dataset$Salary, FUN = function(x) 

                        mean(x, na.rm = TRUE)), 

                      dataset$Salary)

Notice how we used the ave() function here. It takes the average of the specific column you have entered where FUN is a function of x that calculates the mean excluding NA values (na.rm=TRUE).

 else,

 take whatever present in dataset$Age

We’ll use the mean() function now:
#defining  x = 1 2 3

x <- 1:3

#introducing missing value

x[1] <- NA

# mean = NA

mean(x)

## [1] NA

# mean = mean excluding the NA value

mean(x, na.rm = T)

## [1] 2.5

After identifying and fixing the problem, our dataset looks like this:

## Country Age Salary Purchased
## 1 France 44 72000.00 No
## 2 Spain 27 48000.00 Yes
## 3 Germany 30 54000.00 No
## 4 Spain 38 61000.00 No
## 5 Germany 40 63777.78 Yes
## 6 France 35 58000.00 Yes
## 7 Spain 38 52000.00 No
## 8 France 48 79000.00 Yes
## 9 Germany 50 83000.00 No
## 10 France 37 67000.00 Yes

Also Read: Career Opportunities in R Programming Language

2. Step: Categorical Data 

Categorical data is non-numeric data that belongs to particular categories. The Country column in our dataset is categorical data. The read.csv() function in R would make all the string variables as categorical variables. However, we can’t use it in every case. 

Here’s how you can create specific variables as factor variables:
dataset$Country = factor(dataset$Country,

                         levels = c(‘France’, ‘Spain’, ‘Germany’),

                         labels = c(1, 2, 3))

dataset$Purchased = factor(dataset$Purchased,

                           levels = c(‘No’, ‘Yes’),

                           labels = c(0, 1))

3. Step: Splitting Data

Now, we have to split our dataset into two separate datasets. One for training our machine learning model while the other one for testing the same. 

To do so, we’ll first install the caTools package (if not available) and add it to our library. Afterwards, we’ll use the set.seed() function to ensure that the split is done randomly. Use the following code:
library(caTools)

set.seed(123) 

split = sample.split(dataset$Purchased,SplitRatio = 0.8)

training_set = subset(dataset,split == TRUE)

test_set = subset(dataset, split == FALSE)

You must have noticed that we have kept the split ratio as 80:20. This is because it is the most conventional split ratio for training sets and test sets. Our sample.split() method has taken the column and created a numeric array with randomized true and false values according to the split ratio. 

4. Step: Feature Scaling or Overfitting

Feature scaling is required when different features in your dataset have different ranges. In our case, the Age and Salary columns have different ranges, which can cause problems in training our ML model. 

When you have a feature with a significantly higher range than the other feature, the euclidean distance increases considerably, causing the model to give wrong results. 

Note that most libraries in R fix this issue automatically but it’s important to know how to fix this. Do the following:

training_set[,2:3] = scale(training_set[,2:3])

test_set[,2:3] = scale(test_set[,2:3])

It would fix the issue and your training set’s features would have the same ranges, minimizing the chances of any errors during machine learning implementations. 

Conclusion

We hope that you found our data preprocessing in R tutorial helpful. It would be best to understand the tutorial before you try testing it out yourself. Understanding the concepts is much more important than using them. 

What are your thoughts on our data preprocessing in R tutorial? Share them in the comments below.

If you are curious to learn about R, data science, check out our PG Diploma in Data Science which is created for working professionals and offers 10+ case studies & projects, practical hands-on workshops, mentorship with industry experts, 1-on-1 with industry mentors, 400+ hours of learning and job assistance with top firms.

Prepare for a Career of the Future

PG DIPLOMA FROM IIIT-B, 100+ HRS OF CLASSROOM LEARNING, 400+ HRS OF ONLINE LEARNING & 360 DEGREES CAREER SUPPORT
LEARN MORE

Leave a comment

Your email address will not be published.

×