Business Intelligence Assignment 5

Load necessary libraries and datasets

dirty_iris <- read.csv("https://raw.githubusercontent.com/edwindj/datacleaning/master/data/dirty_iris.csv")
library(dplyr)
library(DMwR2)  # For kNN imputation

Question 1

Determine missing values of Petal.length

sum(is.na(dirty_iris$Petal.Length))

## [1] 19

Question 2

Number and the percentage of observations that are complete

num_complete <- sum(complete.cases(dirty_iris))
percentage_complete <- (num_complete / nrow(dirty_iris)) * 100
print(num_complete,) %>% cat(" % \n",percentage_complete,"%")

## [1] 96
## 96  % 
##  64 %

Question 3

Check for special values in numeric columns

num_cols <- sapply(dirty_iris, is.numeric)
has_NA <- any(is.na(dirty_iris[, num_cols])) 
print(has_NA)

## [1] TRUE

has_NAn <- any(is.nan(as.matrix(dirty_iris[, num_cols])))
print(has_NAn)

## [1] FALSE

has_INF <- any(is.infinite(as.matrix(dirty_iris[, num_cols])))
print(has_INF)

## [1] TRUE

Question 4

Locate positions of infinite values and replace infinite values with NA

inf_positions <- which(is.infinite(as.matrix(dirty_iris[, num_cols])), arr.ind = TRUE)

dirty_iris[inf_positions] <- NA

Question 5

Identify rows where Sepal.Width is not positive or Sepal.Length exceeds 30 cm, count the observations.

errors <- which((dirty_iris$Sepal.Width <= 0 | dirty_iris$Sepal.Length > 30) & 
                           !is.na(dirty_iris$Sepal.Width) & !is.na(dirty_iris$Sepal.Length))

# Subset the dataset using the found indices
violations <- dirty_iris[errors, ]
print(violations)

##     Sepal.Length Sepal.Width Petal.Length Petal.Width    Species
## 16           5.0          -3          3.5         1.0 versicolor
## 28          73.0          29         63.0          NA  virginica
## 125         49.0          30         14.0         2.0     setosa
## 130          5.7           0          1.7         0.3     setosa

print(length(errors))

## [1] 4

Question 6

Locate observations where Sepal.Width is ≤ 0

invalid_width <- which(dirty_iris$Sepal.Width <= 0 & !is.na(dirty_iris$Sepal.Width))

dirty_iris$Sepal.Width[invalid_width] <- ifelse(dirty_iris$Sepal.Width[invalid_width] < 0,abs(dirty_iris$Sepal.Width[invalid_width]),NA)
print(dirty_iris[invalid_width, ])

##     Sepal.Length Sepal.Width Petal.Length Petal.Width    Species
## 16           5.0           3          3.5         1.0 versicolor
## 130          5.7          NA          1.7         0.3     setosa

Question 7

Impute Sepal.Width using mean

dirty_iris$Sepal.Width[is.na(dirty_iris$Sepal.Width)] <- mean(dirty_iris$Sepal.Width, na.rm = TRUE)

Impute Petal.Length using median

dirty_iris$Petal.Length[is.na(dirty_iris$Petal.Length)] <- median(dirty_iris$Petal.Length, na.rm = TRUE)

Summary of dirty_iris

summary(dirty_iris)

##   Sepal.Length     Sepal.Width      Petal.Length     Petal.Width   
##  Min.   : 0.000   Min.   : 2.200   Min.   : 0.000   Min.   :0.100  
##  1st Qu.: 5.100   1st Qu.: 2.825   1st Qu.: 1.700   1st Qu.:0.300  
##  Median : 5.750   Median : 3.100   Median : 4.500   Median :1.300  
##  Mean   : 6.559   Mean   : 3.462   Mean   : 4.456   Mean   :1.207  
##  3rd Qu.: 6.400   3rd Qu.: 3.462   3rd Qu.: 5.100   3rd Qu.:1.800  
##  Max.   :73.000   Max.   :30.000   Max.   :63.000   Max.   :2.500  
##  NA's   :10                                         NA's   :13     
##    Species         
##  Length:150        
##  Class :character  
##  Mode  :character  
##                    
##                    
##                    
##