Creating a Data Frame with the Mushrooms Dataset

I downloaded the mushrooms dataset from https://archive.ics.uci.edu/ml/datasets/Mushroom. Then I uploaded the file to my github repository. I used the curl package, to retrieve the dataset from my repository.

library(curl)
y <- read.csv(curl("https://raw.githubusercontent.com/nabilahossain/Image/master/assignment%203.txt"), header = FALSE)
head(y[1:10], 10)

##    V1 V2 V3 V4 V5 V6 V7 V8 V9 V10
## 1   p  x  s  n  t  p  f  c  n   k
## 2   e  x  s  y  t  a  f  c  b   k
## 3   e  b  s  w  t  l  f  c  b   n
## 4   p  x  y  w  t  p  f  c  n   n
## 5   e  x  s  g  f  n  f  w  b   k
## 6   e  x  y  y  t  a  f  c  b   n
## 7   e  b  s  w  t  a  f  c  b   g
## 8   e  b  y  w  t  l  f  c  b   n
## 9   p  x  y  w  t  p  f  c  n   p
## 10  e  b  s  y  t  a  f  c  b   g

I choose to look at the columns that hold information for “classes”, “cap shape”, “cap surface”, “odor”, “population” and “habitat” from the mushrooms dataset. I choose appropriate columns and assign their names.

x <- data.frame(y[c(1, 2, 3, 6, 22, 23)])
names(x) <- c("Classes", "Cap_Shape", "Cap_Surface", "Odor", "Population", "Habitat")
head(x, 3)

##   Classes Cap_Shape Cap_Surface Odor Population Habitat
## 1       p         x           s    p          s       u
## 2       e         x           s    a          n       g
## 3       e         b           s    l          n       m

I check if I got the right number of rows and columns.

ncol(x)

## [1] 6

nrow(x)

## [1] 8124

I will replace the abbreviations for “classes”, and check if they are correct.

x$Classes <- ifelse(x$Classes=="p", "Poisonous", 
              ifelse(x$Classes=="e", "Edible", "N/A")
)
head(x, 5)

##     Classes Cap_Shape Cap_Surface Odor Population Habitat
## 1 Poisonous         x           s    p          s       u
## 2    Edible         x           s    a          n       g
## 3    Edible         b           s    l          n       m
## 4 Poisonous         x           y    p          s       u
## 5    Edible         x           s    n          a       g

Next I will replace the abbreviations for columns: “classes”, “cap_shape”, “cap_surface”, “odor”, “population” and “habitat”.

For Cap_Shape:

x$Cap_Shape <- ifelse(x$Cap_Shape=="b", "Bell", 
              ifelse(x$Cap_Shape=="c", "Conical",
              ifelse(x$Cap_Shape=="x", "Convex",
              ifelse(x$Cap_Shape=="f", "Flat",
              ifelse(x$Cap_Shape=="k", "Knobbed",
              ifelse(x$Cap_Shape=="s", "Sunken", "N/A")
)))))
head(x, 3)

##     Classes Cap_Shape Cap_Surface Odor Population Habitat
## 1 Poisonous    Convex           s    p          s       u
## 2    Edible    Convex           s    a          n       g
## 3    Edible      Bell           s    l          n       m

For Cap_Surface and Odor:

x$Cap_Surface <- ifelse(x$Cap_Surface=="f", "Fibrous",
              ifelse(x$Cap_Surface=="g", "Grooves",
              ifelse(x$Cap_Surface=="y", "Scaly",
              ifelse(x$Cap_Surface=="s", "Smooth", "N/A")
)))
x$Odor <- ifelse(x$Odor=="a", "Almond",  
              ifelse(x$Odor=="l", "Anise",
              ifelse(x$Odor=="c", "Creosote",
              ifelse(x$Odor=="y", "Fishy",
              ifelse(x$Odor=="f", "Foul", 
              ifelse(x$Odor=="m", "Musty",
              ifelse(x$Odor=="n", "None",
              ifelse(x$Odor=="p", "Pungent",
              ifelse(x$Odor=="s", "Spicy", "N/A")
))))))))
tail(x, 3)

##        Classes Cap_Shape Cap_Surface  Odor Population Habitat
## 8122    Edible      Flat      Smooth  None          c       l
## 8123 Poisonous   Knobbed       Scaly Fishy          v       l
## 8124    Edible    Convex      Smooth  None          c       l

For Population and Habitat:

x$Population <- ifelse(x$Population=="a", "Abundant",
              ifelse(x$Population=="c", "Clustered",
              ifelse(x$Population=="n", "Numerous",
              ifelse(x$Population=="s", "Scattered",
              ifelse(x$Population=="v", "Several",
              ifelse(x$Population=="y", "Solitary", "N/A")
)))))

x$Habitat <- ifelse(x$Habitat=="g", "grasses",  
              ifelse(x$Habitat=="l", "Leaves",
              ifelse(x$Habitat=="m", "Meadows",
              ifelse(x$Habitat=="p", "Paths",       
              ifelse(x$Habitat=="u", "Urban",
              ifelse(x$Habitat=="w", "Waste",
              ifelse(x$Habitat=="d", "Woods", "N/A")
))))))

head(x, 3)

##     Classes Cap_Shape Cap_Surface    Odor Population Habitat
## 1 Poisonous    Convex      Smooth Pungent  Scattered   Urban
## 2    Edible    Convex      Smooth  Almond   Numerous grasses
## 3    Edible      Bell      Smooth   Anise   Numerous Meadows

I check if I was able to replace the abbreviations for the subset correctly. I look at the first 30 rows (without unique row number) of the subset of mushrooms in a table format (using the knitr package):

rownames(x) <- NULL
library(knitr)
kable(head(x, 30))

Classes	Cap_Shape	Cap_Surface	Odor	Population	Habitat
Poisonous	Convex	Smooth	Pungent	Scattered	Urban
Edible	Convex	Smooth	Almond	Numerous	grasses
Edible	Bell	Smooth	Anise	Numerous	Meadows
Poisonous	Convex	Scaly	Pungent	Scattered	Urban
Edible	Convex	Smooth	None	Abundant	grasses
Edible	Convex	Scaly	Almond	Numerous	grasses
Edible	Bell	Smooth	Almond	Numerous	Meadows
Edible	Bell	Scaly	Anise	Scattered	Meadows
Poisonous	Convex	Scaly	Pungent	Several	grasses
Edible	Bell	Smooth	Almond	Scattered	Meadows
Edible	Convex	Scaly	Anise	Numerous	grasses
Edible	Convex	Scaly	Almond	Scattered	Meadows
Edible	Bell	Smooth	Almond	Scattered	grasses
Poisonous	Convex	Scaly	Pungent	Several	Urban
Edible	Convex	Fibrous	None	Abundant	grasses
Edible	Sunken	Fibrous	None	Solitary	Urban
Edible	Flat	Fibrous	None	Abundant	grasses
Poisonous	Convex	Smooth	Pungent	Scattered	grasses
Poisonous	Convex	Scaly	Pungent	Scattered	Urban
Poisonous	Convex	Smooth	Pungent	Scattered	Urban
Edible	Bell	Smooth	Almond	Scattered	Meadows
Poisonous	Convex	Scaly	Pungent	Several	grasses
Edible	Bell	Scaly	Anise	Scattered	Meadows
Edible	Bell	Scaly	Almond	Numerous	Meadows
Edible	Bell	Smooth	Anise	Scattered	Meadows
Poisonous	Flat	Smooth	Pungent	Several	grasses
Edible	Convex	Scaly	Almond	Numerous	Meadows
Edible	Convex	Scaly	Anise	Numerous	Meadows
Edible	Flat	Fibrous	None	Solitary	Urban
Edible	Convex	Smooth	Almond	Several	Woods

If we wanted to look at only the first 15 edible mushrooms (without unique row number), in a table format, it would look like this:

SetE <- subset(x, Classes=="Edible")

Classes	Cap_Shape	Cap_Surface	Odor	Population	Habitat
Edible	Convex	Smooth	Almond	Numerous	grasses
Edible	Bell	Smooth	Anise	Numerous	Meadows
Edible	Convex	Smooth	None	Abundant	grasses
Edible	Convex	Scaly	Almond	Numerous	grasses
Edible	Bell	Smooth	Almond	Numerous	Meadows
Edible	Bell	Scaly	Anise	Scattered	Meadows
Edible	Bell	Smooth	Almond	Scattered	Meadows
Edible	Convex	Scaly	Anise	Numerous	grasses
Edible	Convex	Scaly	Almond	Scattered	Meadows
Edible	Bell	Smooth	Almond	Scattered	grasses
Edible	Convex	Fibrous	None	Abundant	grasses
Edible	Sunken	Fibrous	None	Solitary	Urban
Edible	Flat	Fibrous	None	Abundant	grasses
Edible	Bell	Smooth	Almond	Scattered	Meadows
Edible	Bell	Scaly	Anise	Scattered	Meadows

Creating a Data Frame with the Mushrooms Dataset

Nabila Hossain

July 19, 2015