Scribd
Upload
9 views4 pages

Data Mining and R Programming Guide

The document outlines the steps for data preprocessing and k-means clustering in R, including importing data, handling missing values, and normalizing data. It also compares k-means and hierarchical clustering methods, detailing their characteristics and functions in R. Additionally, it describes how to create custom functions and lists various data types available in R, along with methods to check a variable's data type.

Uploaded by

suraj.199586
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
9 views4 pages

Data Mining and R Programming Guide

The document outlines the steps for data preprocessing and k-means clustering in R, including importing data, handling missing values, and normalizing data. It also compares k-means and hierarchical clustering methods, detailing their characteristics and functions in R. Additionally, it describes how to create custom functions and lists various data types available in R, along with methods to check a variable's data type.

Uploaded by

suraj.199586
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Assignment: Data Mining and R Programming

1. Explain the steps to perform data preprocessing in R for a data mining


task.

Answer:

Data preprocessing is an essential step before applying data mining algorithms. The steps in R
are:

a. Import Data:

data <- [Link]("[Link]")

b. Handle Missing Values:

Check missing values:​


​
sum([Link](data))

Replace missing values (e.g., with mean):​


​
data$column[[Link](data$column)] <- mean(data$column, [Link] = TRUE)

c. Remove Duplicates:

data <- unique(data)

d. Data Transformation (e.g., normalization):

data$column <- scale(data$column)

e. Encoding categorical variables:

data$category <- [Link](data$category)

f. Feature Selection/Extraction:

Use correlation or other statistical methods to remove irrelevant features.


2. Discuss the steps involved in performing k-means clustering in R.

Answer:

a. Load and prepare the dataset:

data <- iris[, -5] # Removing label for clustering

b. Normalize the data:

data <- scale(data)

c. Apply k-means algorithm:

[Link](123)

kmeans_result <- kmeans(data, centers = 3)

d. Examine results:

kmeans_result$cluster

e. Visualize clusters:

library(ggplot2)

data_clustered <- [Link](data, cluster = kmeans_result$cluster)

ggplot(data_clustered, aes(x = data[,1], y = data[,2], color =


[Link](cluster))) +

geom_point()

3. Discuss the steps involved in performing k-means clustering in R.

Feature K-means Clustering Hierarchical Clustering

Type Partitional Hierarchical (agglomerative


or divisive)

Predefined clusters Must specify number of No need to specify;


clusters (k) dendrogram helps choose
Scalability Efficient for large datasets Slower for large datasets

Output Cluster assignments Dendrogram showing nested


clusters

Function in R kmeans() hclust() + cutree()

Distance Metric Uses Euclidean distance Allows various metrics


(Euclidean, Manhattan, etc.)

Visualization Basic or 2D plots Dendrogram

4 . How would you create a custom function in R? Provide an example.

Answer:

You can define a custom function in R using the function keyword.

Example:

# Custom function to calculate square of a number

square <- function(x) {

return(x^2)

# Calling the function

square(5)

# Output: 25

5. What are the different data types available in R, and how


can you check the data type of a variable?

Answer:

Common data types in R:


●​ Numeric: Decimal values (e.g., 3.14)​

●​ Integer: Whole numbers (e.g., 5L)​

●​ Character: Text (e.g., "Hello")​

●​ Logical: Boolean (TRUE or FALSE)​

●​ Complex: Complex numbers (e.g., 2 + 3i)​

●​ Factor: Categorical variables​

●​ Raw: Raw bytes​

Check data type of a variable:

class(x) # returns class of the object

typeof(x) # returns the internal type

[Link](x)

[Link](x)

Example:

x <- "Hello"

class(x) # "character"

typeof(x) # "character"

You might also like