Introduction

We would like to clasify characteristics of mushrooms as edible and poisonous using data visualization techniques.

Attribute Information: (classes: edible=e, poisonous=p)

cap-shape: bell=b,conical=c,convex=x,flat=f, knobbed=k,sunken=s
cap-surface: fibrous=f,grooves=g,scaly=y,smooth=s
cap-color: brown=n,buff=b,cinnamon=c,gray=g,green=r,pink=p,purple=u,red=e,white=w,yellow=y
bruises: bruises=t,no=f
odor: almond=a,anise=l,creosote=c,fishy=y,foul=f,musty=m,none=n,pungent=p,spicy=s
gill-attachment: attached=a,descending=d,free=f,notched=n
gill-spacing: close=c,crowded=w,distant=d
gill-size: broad=b,narrow=n
gill-color: black=k,brown=n,buff=b,chocolate=h,gray=g, green=r,orange=o,pink=p,purple=u,red=e,white=w,yellow=y
stalk-shape: enlarging=e,tapering=t
stalk-root: bulbous=b,club=c,cup=u,equal=e,rhizomorphs=z,rooted=r,missing=?
stalk-surface-above-ring: fibrous=f,scaly=y,silky=k,smooth=s
stalk-surface-below-ring: fibrous=f,scaly=y,silky=k,smooth=s
stalk-color-above-ring: brown=n,buff=b,cinnamon=c,gray=g,orange=o,pink=p,red=e,white=w,yellow=y
stalk-color-below-ring: brown=n,buff=b,cinnamon=c,gray=g,orange=o,pink=p,red=e,white=w,yellow=y
veil-type: partial=p,universal=u
veil-color: brown=n,orange=o,white=w,yellow=y
ring-number: none=n,one=o,two=t
ring-type: cobwebby=c,evanescent=e,flaring=f,large=l,none=n,pendant=p,sheathing=s,zone=z
spore-print-color: black=k,brown=n,buff=b,chocolate=h,green=r,orange=o,purple=u,white=w,yellow=y
population: abundant=a,clustered=c,numerous=n,scattered=s,several=v,solitary=y
habitat: grasses=g,leaves=l,meadows=m,paths=p,urban=u,waste=w,woods=d

Load Packages

library(tidyverse)
library(gridExtra) # for multiple graphs
library(DT) 
library(reshape2) # for melt

Load Dataset

data <- read_csv("mushrooms.csv")

datatable(head(data))

data <- as_tibble(lapply(data, function(x) as.factor(x))) # factor

Summary of variables

summary(data)

##  class    cap-shape cap-surface   cap-color    bruises       odor     
##  e:4208   b: 452    f:2320      n      :2284   f:4748   n      :3528  
##  p:3916   c:   4    g:   4      g      :1840   t:3376   f      :2160  
##           f:3152    s:2556      e      :1500            s      : 576  
##           k: 828    y:3244      y      :1072            y      : 576  
##           s:  32                w      :1040            a      : 400  
##           x:3656                b      : 168            l      : 400  
##                                 (Other): 220            (Other): 484  
##  gill-attachment gill-spacing gill-size   gill-color   stalk-shape
##  a: 210          c:6812       b:5612    b      :1728   e:3516     
##  f:7914          w:1312       n:2512    p      :1492   t:4608     
##                                         w      :1202              
##                                         n      :1048              
##                                         g      : 752              
##                                         h      : 732              
##                                         (Other):1170              
##  stalk-root stalk-surface-above-ring stalk-surface-below-ring
##  ?:2480     f: 552                   f: 600                  
##  b:3776     k:2372                   k:2304                  
##  c: 556     s:5176                   s:4936                  
##  e:1120     y:  24                   y: 284                  
##  r: 192                                                      
##                                                              
##                                                              
##  stalk-color-above-ring stalk-color-below-ring veil-type veil-color
##  w      :4464           w      :4384           p:8124    n:  96    
##  p      :1872           p      :1872                     o:  96    
##  g      : 576           g      : 576                     w:7924    
##  n      : 448           n      : 512                     y:   8    
##  b      : 432           b      : 432                               
##  o      : 192           o      : 192                               
##  (Other): 140           (Other): 156                               
##  ring-number ring-type spore-print-color population habitat 
##  n:  36      e:2776    w      :2388      a: 384     d:3148  
##  o:7488      f:  48    n      :1968      c: 340     g:2148  
##  t: 600      l:1296    k      :1872      n: 400     l: 832  
##              n:  36    h      :1632      s:1248     m: 292  
##              p:3968    r      :  72      v:4040     p:1144  
##                        b      :  48      y:1712     u: 368  
##                        (Other): 144                 w: 192

We see that there is just a veil type which is p (partial). We delete it.

data <- select(data, -`veil-type`)

Is there missing value?

lapply(data, function(x) 1 - length(which(is.na(x))) / nrow(data) ) %>%
  as_tibble() %>%
  gather(everything(), key = "column", value ="fill_rate") %>%
  ggplot(aes(x = column, y = fill_rate)) + 
  geom_bar(stat = "identity", fill = "blue") +
  coord_flip()

There is no missing value.

Exploration

How many mushrooms are edible or poisonous?

data %>%
  group_by(class) %>%
  summarise(n = n(), percent = round(n() / nrow(data) * 100))

## # A tibble: 2 x 3
##    class     n percent
##   <fctr> <int>   <dbl>
## 1      e  4208      52
## 2      p  3916      48

How cap effects mushrooms?

Our factors start with cap-

p1 <- ggplot(data, aes(x = `cap-shape`)) + geom_histogram(stat = "count", aes(fill = class), position = "fill")
p2 <- ggplot(data, aes(x = `cap-surface`)) + geom_histogram(stat = "count", aes(fill = class), position = "fill")
p3 <- ggplot(data, aes(x = `cap-color`)) + geom_histogram(stat = "count", aes(fill = class), position = "fill")

grid.arrange(p1,p2,p3, ncol = 1)

when cap-shape is conical, it is also poisonous.
when cap-shape is sunken, it is also edible and most of bell-shaped mushrooms are edible.
if cap-surface is grooves, it is also poisionous. (4 observations.)
if cap-color is green or purple, it is also edible.

Bruises and Odor

p1 <- ggplot(data, aes(x = bruises)) + geom_histogram(stat = "count", aes(fill = class), position = "fill")
p2 <- ggplot(data, aes(x = odor)) + geom_histogram(stat = "count", aes(fill = class), position = "fill")
p3 <- ggplot(data, aes(x = odor)) + geom_histogram(stat = "count", aes(fill = class), position = "dodge")
p4 <- ggplot(data, aes(x = bruises)) + geom_histogram(stat = "count", aes(fill = class), position = "dodge")

grid.arrange(p1,p4,p2,p3, ncol = 2)

We can see that odor is important factor and it directly affects class of mushrooms.

How gill affects class?

p1 <- ggplot(data, aes(x = `gill-attachment`)) + geom_histogram(stat = "count", aes(fill = class), position = "fill")
p2 <- ggplot(data, aes(x = `gill-spacing`)) + geom_histogram(stat = "count", aes(fill = class), position = "fill")
p3 <- ggplot(data, aes(x = `gill-size`)) + geom_histogram(stat = "count", aes(fill = class), position = "fill")
p4 <- ggplot(data, aes(x = `gill-color`)) + geom_histogram(stat = "count", aes(fill = class), position = "fill")

grid.arrange(p1,p2,p3,p4, ncol = 2)

We have to see also position with “dodge”

p1 <- ggplot(data, aes(x = `gill-attachment`)) + geom_histogram(stat = "count", aes(fill = class), position = "dodge")
p2 <- ggplot(data, aes(x = `gill-spacing`)) + geom_histogram(stat = "count", aes(fill = class), position = "dodge")
p3 <- ggplot(data, aes(x = `gill-size`)) + geom_histogram(stat = "count", aes(fill = class), position = "dodge")
p4 <- ggplot(data, aes(x = `gill-color`)) + geom_histogram(stat = "count", aes(fill = class), position = "dodge")

grid.arrange(p1,p2,p3,p4, ncol = 2)

gill-color is important factor.

How stalk affects class ?

p1 <- ggplot(data, aes(x = `stalk-surface-below-ring`)) + geom_histogram(stat = "count", aes(fill = class), position = "fill")
p2 <- ggplot(data, aes(x = `stalk-surface-above-ring`)) + geom_histogram(stat = "count", aes(fill = class), position = "fill")
p3 <- ggplot(data, aes(x = `stalk-shape`)) + geom_histogram(stat = "count", aes(fill = class), position = "fill")
p4 <- ggplot(data, aes(x = `stalk-root`)) + geom_histogram(stat = "count", aes(fill = class), position = "fill")
p5 <- ggplot(data, aes(x = `stalk-color-below-ring`)) + geom_histogram(stat = "count", aes(fill = class), position = "fill")
p6 <- ggplot(data, aes(x = `stalk-color-above-ring`)) + geom_histogram(stat = "count", aes(fill = class), position = "fill")

grid.arrange(p1,p2,p3,p4,p5,p6, ncol = 2)

p1 <- ggplot(data, aes(x = `stalk-surface-below-ring`)) + geom_histogram(stat = "count", aes(fill = class), position = "dodge")
p2 <- ggplot(data, aes(x = `stalk-surface-above-ring`)) + geom_histogram(stat = "count", aes(fill = class), position = "dodge")
p3 <- ggplot(data, aes(x = `stalk-shape`)) + geom_histogram(stat = "count", aes(fill = class), position = "dodge")
p4 <- ggplot(data, aes(x = `stalk-root`)) + geom_histogram(stat = "count", aes(fill = class), position = "dodge")
p5 <- ggplot(data, aes(x = `stalk-color-below-ring`)) + geom_histogram(stat = "count", aes(fill = class), position = "dodge")
p6 <- ggplot(data, aes(x = `stalk-color-above-ring`)) + geom_histogram(stat = "count", aes(fill = class), position = "dodge")

grid.arrange(p1,p2,p3,p4,p5,p6, ncol = 2)

We see that properties of below and above are same. We don’t need same variables.

length(which(data$`stalk-color-above-ring` == data$`stalk-color-below-ring`)) / nrow(data) # for color

## [1] 0.6238306

length(which(data$`stalk-surface-above-ring` == data$`stalk-surface-below-ring`)) / nrow(data) # for surface

## [1] 0.770064

stalk-color-below/above-ring is important factor for class.

spore-print-color

p1 <- ggplot(data, aes(x = `spore-print-color`)) + geom_histogram(stat = "count", aes(fill = class), position = "dodge")
p2 <- ggplot(data, aes(x = `spore-print-color`)) + geom_histogram(stat = "count", aes(fill = class), position = "fill")

grid.arrange(p1,p2, ncol = 2)

We see easily that spore-print-color is important factor.

How ring affects class?

p1 <- ggplot(data, aes(x = `ring-type`)) + geom_histogram(stat = "count", aes(fill = class), position = "dodge")
p2 <- ggplot(data, aes(x = `ring-type`)) + geom_histogram(stat = "count", aes(fill = class), position = "fill")
p3 <- ggplot(data, aes(x = `ring-number`)) + geom_histogram(stat = "count", aes(fill = class), position = "dodge")
p4 <- ggplot(data, aes(x = `ring-number`)) + geom_histogram(stat = "count", aes(fill = class), position = "fill")

grid.arrange(p1,p2,p3,p4, ncol = 2)

ring-type can be put on the model.

veil-color

p1 <- ggplot(data, aes(x = `veil-color`)) + geom_histogram(stat = "count", aes(fill = class), position = "dodge")
p2 <- ggplot(data, aes(x = `veil-color`)) + geom_histogram(stat = "count", aes(fill = class), position = "fill")

grid.arrange(p1,p2, ncol = 2)

Population and Habitat

p1 <- ggplot(data, aes(x = `population`)) + geom_histogram(stat = "count", aes(fill = class), position = "dodge")
p2 <- ggplot(data, aes(x = `population`)) + geom_histogram(stat = "count", aes(fill = class), position = "fill")
p3 <- ggplot(data, aes(x = `habitat`)) + geom_histogram(stat = "count", aes(fill = class), position = "dodge")
p4 <- ggplot(data, aes(x = `habitat`)) + geom_histogram(stat = "count", aes(fill = class), position = "fill")

grid.arrange(p1,p2,p3,p4, ncol = 2)

We can say that population and habitat are important for classification.

Parallel Coordinates

library(parcoords)
library(htmlwidgets)
data %>%
  select(class,odor, `gill-color`, `stalk-color-below-ring`, `spore-print-color`, habitat, population) %>%
  parcoords(brushMode = "1D-axes",
          reorderable = TRUE,
          rownames = FALSE,
          color = list(
            colorBy = "class",
            colorScale = htmlwidgets::JS("d3.scale.category10()")
          )    
  )

Results

if mushroom doesn’t have odor, it’s likely edible and if mushroom has smell of anise and almond, it’s certainly edible. In other cases, mushrooms are poisonous.
if mushroom has black-gill, it’s poisonous.
if spore-print-color is green, it’s poisionous and then if it is buff, yellow, orrange or purple, it’s edible.

Mushrooms

Sercan Doğan

10 December 2017