An Introduction to E1071: The Machine Learning Package in R

Machine learning is a powerful tool that allows us to analyze and make predictions based on data. It is a subset of artificial intelligence (AI) that focuses on the development of algorithms that can learn from and make predictions on data. There are many different approaches to machine learning, and one popular tool for implementing machine learning techniques in R is the e1071 package.

In this tutorial, we will introduce the e1071 package and demonstrate how to use it for machine learning in R.

Installation

First, let’s install the e1071 package. Open up your R console and type the following command:

1
install.packages("e1071")

Next, load the package by running the following command:

1
library(e1071)

Data Preparation

Before we can start using the e1071 package for machine learning, we need to have some data to work with. In this tutorial, we will use the iris dataset, which is included with the e1071 package.

To load the iris dataset, run the following command:

1
data(iris)

This will load the iris dataset into your R environment. You can view the dataset by running the following command:

1
head(iris)

This will display the first six rows of the iris dataset. The dataset consists of 150 rows and 5 columns, where the first four columns represent the features (sepal length, sepal width, petal length, and petal width) and the last column represents the species of iris.

Before we can use the iris dataset for machine learning, we need to split it into a training set and a test set. We can do this using the sample.split function from the caTools package.

First, install the caTools package by running the following command:

1
install.packages("caTools")

Next, load the caTools package and split the iris dataset into a training set and a test set by running the following commands:

1
library(caTools)
2

3
split = sample.split(iris$Species, SplitRatio = 0.75)
4
train = subset(iris, split == TRUE)
5
test = subset(iris, split == FALSE)

Model Training

Now that we have our data prepared, we can start training a machine learning model using the e1071 package.

One popular classification algorithm in the e1071 package is the support vector machine (SVM). To train an SVM model on the iris dataset, we can use the svm function.

First, let’s train a linear SVM model by running the following command:

1
model = svm(Species ~ ., data = train)

This command trains a linear SVM model on the training set and stores the model in the model variable.

We can also train a non-linear SVM model by specifying the kernel parameter. For example, to train a radial basis function (RBF) SVM model, we can run the following command:

1
model = svm(Species ~ ., data = train, kernel = "radial")

Model Evaluation

Now that we have trained a machine learning model, we can evaluate its performance on the test set.

To evaluate the performance of the model, we can use the predict function. This function takes in the trained model and the test set, and returns the predicted labels for the test set.

To use the predict function, run the following command:

1
predictions = predict(model, test)

This will store the predicted labels for the test set in the predictions variable.

To measure the accuracy of the model, we can compare the predicted labels to the true labels of the test set. One way to do this is to use the confusionMatrix function from the caret package.

First, install the caret package by running the following command:

1
install.packages("caret")

Next, load the caret package and use the confusionMatrix function by running the following commands:

1
library(caret)
2

3
confusionMatrix(predictions, test$Species)

This will display a confusion matrix showing the number of true positive, true negative, false positive, and false negative predictions made by the model.

Conclusion

In this tutorial, we introduced the e1071 package for machine learning in R and demonstrated how to use it to train and evaluate a support vector machine model on the iris dataset. We also saw how to use the confusionMatrix function to measure the accuracy of the model.

There are many other machine learning algorithms and techniques available in the e1071 package, as well as other packages in R. We encourage you to explore these resources and continue learning about machine learning in R.