Explain with a case study how to implement unsupervised learning algorithm using R.

For unsupervised learning, we will be performing clustering algorithms such as K-Means algorithm.

First let us read the data

crime <- read.csv(file.choose())

Now we will normalize the data using scale function

# Normalizing continuous columns to bring them under same scale

normalized_data<-scale(crime[,2:5]) #excluding the X name column before normalizing

View(normalized_data)

wss = NULL

Now we will create a function to create K-Means algorithm

twss <- NULL

for (i in 2:15){

  twss <- c(twss,kmeans(normalized_data,i)$tot.withinss)

}

Now we will plot the clusters and the value of k will be decided on the elbow spotted on the plot

plot(2:15, twss,type="b", xlab="Number of Clusters", ylab="Within groups sum of squares") # Look for an "elbow" in the scree plot #

title(sub = "K-Means Clustering Scree-Plot")

Now we will select the k value from the scree plot

k_3 <- kmeans(normalized_data,3)

str(k_3)

clust=k_3$cluster

Now we will create the aggregates of clusters and put into a dataframe.

final=data.frame(crime,clust)

aggregate(crime[,-1],by=list(final$clust),mean)