Lab-2021-11-09

Question 1

Required Packages

# install.packages("e1071")

Question 2 & 3

Classifaction using Naive Bayes

library("e1071")
df = iris
model = naiveBayes(Species~.,data=df)



predicted_values = predict(model,df[,1:4])


correctly_predicted = sum(predicted_values == df[,"Species"])
print(paste("Correctly predicted",correctly_predicted))

## [1] "Correctly predicted 144"

accuracy = correctly_predicted / nrow(df)
print(paste("accuracy",accuracy))

## [1] "accuracy 0.96"

Question 4

Train test spliting.

library("e1071")
df = iris
train_test_split_percentage = 0.66

train_rows = sample(nrow(df), nrow(df)*train_test_split_percentage)

train_data = df[train_rows,]
test_data = df[-train_rows,]

model = naiveBayes(Species~.,data=train_data)
predicted_values_test = predict(model,test_data[,1:4])


correctly_predicted_test = sum(predicted_values_test == test_data[,"Species"])
print(paste("Correctly predicted on TEST",correctly_predicted_test))

## [1] "Correctly predicted on TEST 48"

accuracy = correctly_predicted_test / nrow(test_data)
print(paste("accuracy on Test Dataset",accuracy))

## [1] "accuracy on Test Dataset 0.941176470588235"

Question 5

Installing the rpart package

# install.packages("rpart")

Question 6, 7 & 8.

Decision trees in R

library("e1071")
library("rpart")
df = iris
model = rpart(Species~.,data=df)



predicted_values = predict(model,df[,1:4], type="class")


correctly_predicted = sum(predicted_values == df[,"Species"])
print(paste("Correctly predicted",correctly_predicted))

## [1] "Correctly predicted 144"

accuracy = correctly_predicted / nrow(df)
print(paste("accuracy",accuracy))

## [1] "accuracy 0.96"

Question 9.

library("e1071")
library("rpart")
df = iris
train_test_split_percentage = 0.66

train_rows = sample(nrow(df), nrow(df)*train_test_split_percentage)

train_data = df[train_rows,]
test_data = df[-train_rows,]

model = naiveBayes(Species~.,data=train_data)
predicted_values_test = predict(model,test_data[,1:4])


correctly_predicted_test = sum(predicted_values_test == test_data[,"Species"])
print(paste("Correctly predicted on TEST",correctly_predicted_test))

## [1] "Correctly predicted on TEST 48"

accuracy = correctly_predicted_test / nrow(test_data)
print(paste("accuracy on Test Dataset",accuracy))

## [1] "accuracy on Test Dataset 0.941176470588235"