GitHub Repository: debakarr/machinelearning
Path: blob/master/Part 3 - Classification/Naive Bayes/[R] Naive Bayes.ipynb
¹⁰⁰⁹ views

Kernel: R

Naive Bayes

Data preprocessing

In [1]:

# Importing the dataset
dataset = read.csv('Social_Network_Ads.csv')
dataset = dataset[3:5]

In [2]:

head(dataset, 10)

Out[2]:

In [3]:

# Encoding the target feature as factor
dataset$Purchased = factor(dataset$Purchased, levels = c(0, 1))

In [4]:

# Splitting the dataset into the Training set and Test set
# install.packages('caTools')
library(caTools)
set.seed(1234)
split = sample.split(dataset$Purchased, SplitRatio = 0.80)
training_set = subset(dataset, split == TRUE)
test_set = subset(dataset, split == FALSE)

In [5]:

head(training_set, 10)

Out[5]:

In [6]:

head(test_set, 10)

Out[6]:

In [7]:

# Feature Scaling
training_set[-3] = scale(training_set[-3])
test_set[-3] = scale(test_set[-3])

In [8]:

head(training_set, 10)

Out[8]:

In [9]:

head(test_set, 10)

Out[9]:

Fitting Naive Bayes classifier to the Training set

In [10]:

library(e1071)
classifier = naiveBayes(x = training_set[-3],
                        y = training_set$Purchased)

Predicting the Test set results

In [11]:

y_pred = predict(classifier, newdata = test_set[-3])

In [12]:

head(y_pred, 10)

Out[12]:

In [13]:

head(test_set[3], 10)

Out[13]:

Making the Confusion Matrix

In [14]:

cm = table(test_set[, 3], y_pred)
cm

Out[14]:

   y_pred
1
48  3
6 23

classifier made 48 + 23 = 71 correct prediction and 6 + 3 = 9 incoreect predictions.

Visualising the Training set results

In [15]:

library(ElemStatLearn)
set = training_set
X1 = seq(min(set[, 1]) - 1, max(set[, 1]) + 1, by = 0.01)
X2 = seq(min(set[, 2]) - 1, max(set[, 2]) + 1, by = 0.01)
grid_set = expand.grid(X1, X2)
colnames(grid_set) = c('Age', 'EstimatedSalary')
y_grid = predict(classifier, newdata = grid_set)
plot(set[, -3],
     main = 'Naive Bayes (Training set)',
     xlab = 'Age', ylab = 'Estimated Salary',
     xlim = range(X1), ylim = range(X2))
contour(X1, X2, matrix(as.numeric(y_grid), length(X1), length(X2)), add = TRUE)
points(grid_set, pch = '.', col = ifelse(y_grid == 1, 'springgreen3', 'tomato'))
points(set, pch = 21, bg = ifelse(set[, 3] == 1, 'green4', 'red3'), col='white')
legend("topright", legend = c("0", "1"), pch = 16, col = c('red3', 'green4'))

Out[15]:

Visualising the Test set results

In [16]:

library(ElemStatLearn)
set = test_set
X1 = seq(min(set[, 1]) - 1, max(set[, 1]) + 1, by = 0.01)
X2 = seq(min(set[, 2]) - 1, max(set[, 2]) + 1, by = 0.01)
grid_set = expand.grid(X1, X2)
colnames(grid_set) = c('Age', 'EstimatedSalary')
y_grid = predict(classifier, newdata = grid_set)
plot(set[, -3], main = 'Naive Bayes (Test set)',
     xlab = 'Age', ylab = 'Estimated Salary',
     xlim = range(X1), ylim = range(X2))
contour(X1, X2, matrix(as.numeric(y_grid), length(X1), length(X2)), add = TRUE)
points(grid_set, pch = '.', col = ifelse(y_grid == 1, 'springgreen3', 'tomato'))
points(set, pch = 21, bg = ifelse(set[, 3] == 1, 'green4', 'red3'), col='white')
legend("topright", legend = c("0", "1"), pch = 16, col = c('red3', 'green4'))

Out[16]:

Naive Bayes is another non-linear classifier. It works on the principle of Bayes theorem.

Here we can see that we got a pretty good classifier with very fewer incorrect predictions. Though no classifier is 100% correct.

Every classifier is wrong , but some are useful.