Assignment_4

#Q3

dirty_iris <- read.csv("https://raw.githubusercontent.com/edwindj/datacleaning/master/data/dirty_iris.csv")

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

## [1] 0

#Q4

sum(complete.cases(dirty_iris))

## [1] 96

mean(complete.cases(dirty_iris)) * 100

## [1] 64

#Q5

num_cols <- sapply(dirty_iris, is.numeric)

sapply(dirty_iris[, num_cols], function(x) {
  c(
    NA_count  = sum(is.na(x)),
    NaN_count = sum(is.nan(x)),
    Inf_count = sum(is.infinite(x))
  )
})

##           Sepal.Length Sepal.Width Petal.Length Petal.Width
## NA_count            10          17           19          12
## NaN_count            0           0            0           0
## Inf_count            0           0            0           1

#Q6

dirty_iris[, num_cols][is.nan(as.matrix(dirty_iris[, num_cols]))] <- NA

dirty_iris[, num_cols][is.infinite(as.matrix(dirty_iris[, num_cols]))] <- NA

sapply(dirty_iris[, num_cols], function(x) {
  c(
    NA_count = sum(is.na(x)),
    NaN_count = sum(is.nan(x)),
    Inf_count = sum(is.infinite(x))
  )
})

##           Sepal.Length Sepal.Width Petal.Length Petal.Width
## NA_count            10          17           19          13
## NaN_count            0           0            0           0
## Inf_count            0           0            0           0

#Q7

violations <- subset(dirty_iris, Sepal.Width <= 0 | Sepal.Length > 30)

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

nrow(violations)

## [1] 4

#Q8

dirty_iris$Sepal.Width[which(dirty_iris$Sepal.Width < 0)] <- 
  abs(dirty_iris$Sepal.Width[which(dirty_iris$Sepal.Width < 0)])

dirty_iris$Sepal.Width[which(dirty_iris$Sepal.Width == 0)] <- NA

dirty_iris[which(dirty_iris$Sepal.Width <= 0 | is.na(dirty_iris$Sepal.Width)), "Sepal.With"]

## NULL

#Q9

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

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

# --- 3. Sepal.Length: linear regression imputation ---
lm_model <- lm(Sepal.Length ~ Sepal.Width + Petal.Length + Petal.Width + Species,
               data = dirty_iris, na.action = na.exclude)
predicted <- predict(lm_model, newdata = dirty_iris)
dirty_iris$Sepal.Length[is.na(dirty_iris$Sepal.Length)] <- 
  predicted[is.na(dirty_iris$Sepal.Length)]

# --- 4. Petal.Width: kNN imputation ---

colSums(is.na(dirty_iris))

## Sepal.Length  Sepal.Width Petal.Length  Petal.Width      Species 
##            0            0            0           13            0