WEEK – 5

a)How to find a correlation matrix and plot the correlation on iris data set

install.packages("corrplot") # If not already installed

library(corrplot)

# Calculate and plot the correlation matrix

cor_matrix <- cor(iris[, 1:4]) # Correlation for numeric columns

corrplot(cor_matrix, method = "circle", main = "Correlation Matrix of Iris Dataset")

b) Plot the correlation plot on dataset and visualize giving an overview of on iris
data.

corrplot(cor_matrix, method = "circle",

main = "Correlation Matrix of Iris Dataset",

tl.col = "black", tl.srt = 45)

c) Analysis of covariance: variance (ANOVA), if data have categorical variables
on iris data. SOURCE CODE:

library(ggplot2)

# Scatter plot

ggplot(iris, aes(x = Sepal.Length, y = Sepal.Width, color = Species)) + geom_point()

# Box plot

ggplot(iris, aes(x = Species, y = Sepal.Length, fill = Species)) + geom_boxplot()

# Correlation heatmap

cor_matrix <- cor(iris[, 1:4])

cor_data <- as.data.frame(as.table(cor_matrix))

ggplot(cor_data, aes(Var1, Var2, fill = Freq)) + geom_tile() + scale_fill_gradient2()

 WEEK – 6

Create Relationship Model & get the Coefficients

# Reduced data points (3 points each)

x <- c(151, 174, 138)

y <- c(63, 81, 56)

# Apply the lm() function

relation <- lm(y ~ x)

# Print the model

print(relation)
Call:

lm(formula = y ~ x)

Coefficients:

(Intercept) x

-42.0787 0.7046

Get the Summary of the Relationship

# Reduced data points (3 points each)

x <- c(151, 174, 138)

y <- c(63, 81, 56)

# Apply the lm() function

relation <- lm(y ~ x)

# Print the summary of the model

print(summary(relation))
Call:

lm(formula = y ~ x)

Residuals:

1 2 3

-1.3180 0.4759 0.8420

Coefficients:

Estimate Std. Error t value Pr(>|t|)

 (Intercept) -42.07874 9.83170 -4.28 0.1461

x 0.70461 0.06341 11.11 0.0571 .

---

Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 1.635 on 1 degrees of freedom

Multiple R-squared: 0.992, Adjusted R-squared: 0.9839

F-statistic: 123.5 on 1 and 1 DF, p-value: 0.05714

Predict the weight of new persons

# Reduced data points (3 points each)

x <- c(151, 174, 138)

y <- c(63, 81, 56)

# Apply the lm() function

relation <- lm(y ~ x)

# Print the summary of the model

print(summary(relation))
Call:

lm(formula = y ~ x)

Residuals:

1 2 3

-1.3180 0.4759 0.8420

Coefficients:

Estimate Std. Error t value Pr(>|t|)

(Intercept) -42.07874 9.83170 -4.28 0.1461

x 0.70461 0.06341 11.11 0.0571 .

---

Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 1.635 on 1 degrees of freedom

Multiple R-squared: 0.992, Adjusted R-squared: 0.9839

F-statistic: 123.5 on 1 and 1 DF, p-value: 0.05714

# Predict y values based on the model

predicted_y <- predict(relation, newdata = data.frame(x = c(160, 145, 170)))

# Print the predicted values

print(predicted_y)
1 2 3

70.65948 60.09027 77.70562

Visualize the Regression Graphically

# Reduced data points (3 points each)

x <- c(151, 174, 138)

y <- c(63, 81, 56)

# Apply the lm() function

relation <- lm(y ~ x)

# Plot the data points

plot(x, y, main = "Height vs Weight Regression",

xlab = "Weight (kg)", ylab = "Height (cm)",

pch = 16, col = "blue")

# Add the regression line

abline(relation, col = "red")

 WEEK – 7

# Example dataset

data(mtcars)

# Convert 'cyl' (cylinder) into a factor (categorical variable)

mtcars$cyl <- factor(mtcars$cyl)

# Apply logistic regression

mylogit <- glm(am ~ mpg + wt + cyl, data = mtcars, family = "binomial")

# Print the summary of the logistic regression model

summary(mylogit)
Call:

glm(formula = am ~ mpg + wt + cyl, family = "binomial", data = mtcars)

Coefficients:

Estimate Std. Error z value Pr(>|z|)

(Intercept) 23.92836 14.17738 1.688 0.0915 .

mpg -0.09851 0.35135 -0.280 0.7792

wt -8.17801 3.34965 -2.441 0.0146 *

cyl6 3.00979 2.51067 1.199 0.2306

cyl8 4.98194 3.50934 1.420 0.1557

---

Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

(Dispersion parameter for binomial family taken to be 1)

Null deviance: 43.230 on 31 degrees of freedom

Residual deviance: 14.588 on 27 degrees of freedom

AIC: 24.588

Number of Fisher Scoring iterations: 7

 WEEK – 8

# Load the mtcars dataset

data(mtcars)

# Convert 'cyl' to a factor

mtcars$cyl <- factor(mtcars$cyl)

# Fit a linear regression model with 'cyl' as the predictor

model <- lm(mpg ~ cyl, data = mtcars)

# Display regression coefficients

coef(summary(model))
Estimate Std. Error t value Pr(>|t|)

(Intercept) 26.663636 0.9718008 27.437347 2.688358e-22

cyl6 -6.920779 1.5583482 -4.441099 1.194696e-04

cyl8 -11.563636 1.2986235 -8.904534 8.568209e-10

# Perform ANOVA on the model

anova(model)
Analysis of Variance Table

Response: mpg

Df Sum Sq Mean Sq F value Pr(>F)

cyl 2 824.78 412.39 39.697 4.979e-09 ***

Residuals 29 301.26 10.39

---

Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
 WEEK – 9

Install relevant pacakages

install.packages("rpart.plot")

install.packages("tree")

install.packages("ISLR")

install.packages("rattle")

library(tree)

library(ISLR)

library(rpart.plot)

library(rattle)

# Load data and clean NA values

Hitters <- na.omit(Hitters)

# Log transform Salary

Hitters$Salary <- log(Hitters$Salary)

# Plot transformed Salary

hist(Hitters$Salary, main = "Log-Transformed Salary", xlab = "Log(Salary)")

# Fit regression tree

tree.fit <- tree(Salary ~ Hits + Years, data = Hitters)

summary(tree.fit)
Regression tree:

tree(formula = Salary ~ Hits + Years, data = Hitters)

Number of terminal nodes: 7

Residual mean deviance: 0.002708 = 0.6933 / 256

Distribution of residuals:

Min. 1st Qu. Median Mean 3rd Qu. Max.

-0.2355000 -0.0258400 -0.0005869 0.0000000 0.0332000 0.2069000

# Plot the tree

plot(tree.fit, uniform = TRUE, margin = 0.2)

text(tree.fit, use.n = TRUE, cex = 0.8)

# Split data into training and testing

set.seed(123)

split <- createDataPartition(Hitters$Salary, p = 0.5, list = FALSE)

train <- Hitters[split, ]

test <- Hitters[-split, ]

# Train a tree model on the training data

trees <- tree(Salary ~ ., data = train)

plot(trees)

text(trees, pretty = 0)

# Cross-validation to prune the tree

cv.trees <- cv.tree(trees)

plot(cv.trees)

# Prune tree to best model

prune.trees <- prune.tree(trees, best = 4)

plot(prune.trees)

text(prune.trees, pretty = 0)

# Predict on test data using the pruned tree

yhat <- predict(prune.trees, test)

# Plot predicted vs actual salary

plot(yhat, test$Salary)
abline(0, 1)

# Calculate mean squared error

mean_squared_error <- mean((yhat - test$Salary)^2)

cat("Mean Squared Error:", mean_squared_error, "\n")

Mean Squared Error: 0.002714588
 WEEK – 10

Clustering algorithms for unsupervised classification.

# Load the cluster library

library(cluster)

# Set random seed for reproducibility

set.seed(20)

# Apply K-means clustering on Sepal.Width and Petal.Width (columns 3 and 4)

irisCluster <- kmeans(iris[, 3:4], centers = 3, nstart = 20)

# Output the clustering result

print(irisCluster)
K-means clustering with 3 clusters of sizes 52, 48, 50

Cluster means:

Petal.Length Petal.Width

1 4.269231 1.342308

2 5.595833 2.037500

3 1.462000 0.246000

Clustering vector:

[1] 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3

[46] 3 3 3 3 3 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 1 1 1 1 1 2 1 1 1 1 1 1

[91] 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 1 2 2 2 2 2 2 2 2 2 2 2 2 1 2 2 2 2 2 2 1 2 2 2 2 2 2 2 2

[136] 2 2 2 1 2 2 2 2 2 2 2 2 2 2 2

Within cluster sum of squares by cluster:

[1] 13.05769 16.29167 2.02200

(between_SS / total_SS = 94.3 %)

Available components:
[1] "cluster" "centers" "totss" "withinss" "tot.withinss" "betweenss"

[7] "size" "iter" "ifault"

# Convert cluster assignments to factor

irisCluster$cluster <- as.factor(irisCluster$cluster)

# Plot the clustering results

library(ggplot2)

ggplot(iris, aes(Petal.Length, Petal.Width, color = irisCluster$cluster)) +

geom_point() +

labs(title = "K-means Clustering on Iris Dataset",

x = "Petal Length", y = "Petal Width")

# Create a distance matrix from the mtcars dataset

d <- dist(as.matrix(mtcars))

# Apply hierarchical clustering

hc <- hclust(d)

# Plot the dendrogram

plot(hc, main = "Hierarchical Clustering of mtcars", xlab = "", sub = "")

# Generate a synthetic dataset with two clusters

x <- rbind(cbind(rnorm(10, 0, 0.5), rnorm(10, 0, 0.5)),

cbind(rnorm(15, 5, 0.5), rnorm(15, 5, 0.5)))

# Apply PAM clustering with 2 clusters and plot the result

library(cluster)

clusplot(pam(x, 2))

# Add noise (25 random values) to the dataset

x4 <- cbind(x, rnorm(25), rnorm(25))

# Apply PAM clustering with 2 clusters and plot the result

library(cluster)

clusplot(pam(x4, 2))