Your task is to study the Mushrooms Dataset in the UCI repository and the associated description of the data (i.e. “data dictionary”). You may need to look around a bit, but it’s there! You should take the data, and create a data frame with a subset of the columns (and if you like rows) in the dataset. You should include the column that indicates edible or poisonous and three or four other columns. You should also add meaningful column names and replace the abbreviations used in the data-for example, in the appropriate column, “e” might become “edible.” Your deliverable is the R code to perform these transformation tasks.
I started with reading the data file into a dataframe. The Header argument needs to be set to false, because this file does not have a header row. Setting na.strings to “?” turns the question marks in column 12 to NA’s.
In most cases I would want a complete copy of any data I was examining in a format that would be the most useful to me. I am still learning the advantges of having Factors in a dataframe vesus just vectors. It seems to me that having the levels for these encoded columns would be useful over the long haul, because it would serve as a built in data dictionary. This is why I did not use stringAsFactors, as.is, or colclasses during the initial read.
df <- read.csv("c:\\Users\\Robert\\Documents\\CUNY\\Bridge Classes\\R Programming\\Week3\\agaricus-lepiota.data.txt", header = FALSE, na.strings = "?")
summary(df)## V1 V2 V3 V4 V5 V6
## 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
## V7 V8 V9 V10 V11 V12 V13
## a: 210 c:6812 b:5612 b :1728 e:3516 b :3776 f: 552
## f:7914 w:1312 n:2512 p :1492 t:4608 c : 556 k:2372
## w :1202 e :1120 s:5176
## n :1048 r : 192 y: 24
## g : 752 NA's:2480
## h : 732
## (Other):1170
## V14 V15 V16 V17 V18 V19
## f: 600 w :4464 w :4384 p:8124 n: 96 n: 36
## k:2304 p :1872 p :1872 o: 96 o:7488
## s:4936 g : 576 g : 576 w:7924 t: 600
## y: 284 n : 448 n : 512 y: 8
## b : 432 b : 432
## o : 192 o : 192
## (Other): 140 (Other): 156
## V20 V21 V22 V23
## e:2776 w :2388 a: 384 d:3148
## f: 48 n :1968 c: 340 g:2148
## 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: 192For this assignment I could subset the dataframe first and only name the columns I plan to keep, however my normal course of action would be to complete the dataset. In work, I might find out I need more columns later and not want to go back to configure them. Here are all the column names assigned to their columns.
fields <- c("edible-poison", "cap-shape", "cap-surface", "cap-color", "bruises", "odor", "gill-attachment", "gill-spacing", "gill-size", "gill-color", "stalk-shape", "stalk-root", "stalk-surface-above-ring", "stalk-surface-below-ring", "stalk-color-above-ring", "stalk-color-below-ring", "veil-type", "veil-color", "ring-number", "ring-type", "spore-print-color", "population", "habitat")
colnames(df) <- fields
head(df)## edible-poison cap-shape cap-surface cap-color bruises odor
## 1 p x s n t p
## 2 e x s y t a
## 3 e b s w t l
## 4 p x y w t p
## 5 e x s g f n
## 6 e x y y t a
## gill-attachment gill-spacing gill-size gill-color stalk-shape stalk-root
## 1 f c n k e e
## 2 f c b k e c
## 3 f c b n e c
## 4 f c n n e e
## 5 f w b k t e
## 6 f c b n e c
## stalk-surface-above-ring stalk-surface-below-ring stalk-color-above-ring
## 1 s s w
## 2 s s w
## 3 s s w
## 4 s s w
## 5 s s w
## 6 s s w
## stalk-color-below-ring veil-type veil-color ring-number ring-type
## 1 w p w o p
## 2 w p w o p
## 3 w p w o p
## 4 w p w o p
## 5 w p w o e
## 6 w p w o p
## spore-print-color population habitat
## 1 k s u
## 2 n n g
## 3 n n m
## 4 k s u
## 5 n a g
## 6 k n gI could apply the same thinking to the field contents and in most cases that is probably what I would do to complete the dataset. In this case there are 23 fileds with some having as many as 10 different possible values. For the assignment I decided that discretion is the better part of valor and subsetted first, before replacing the variables.
small.df <- subset(df, select = c("edible-poison", "cap-shape", "cap-surface", "bruises", "odor"))
head(small.df)## edible-poison cap-shape cap-surface bruises odor
## 1 p x s t p
## 2 e x s t a
## 3 e b s t l
## 4 p x y t p
## 5 e x s f n
## 6 e x y t aTo replace “code” letters with words in each column requires two steps. We need to add the word values to the elements for the factor and then we can replace the letter with more meaningful full word. The levels step could be avoided, if we made the columns vectors and not factors. I need to figure out if the extra work is worth it most times.
# Column 1 field values, add levels to factor before adding values
levels(small.df$'edible-poison') <- c(levels(small.df$'edible-poison'), "edible", "poison")
small.df$'edible-poison'[small.df$'edible-poison' == "e"] <- "edible"
small.df$'edible-poison'[small.df$'edible-poison' == "p"] <- "poison"
# Column 2 field values
levels(small.df$'cap-shape') <- c(levels(small.df$'cap-shape'), "bell", "conical", "convex", "flat", "knobbed", "sunken")
small.df$'cap-shape'[small.df$'cap-shape' == "b"] <- "bell"
small.df$'cap-shape'[small.df$'cap-shape' == "c"] <- "conical"
small.df$'cap-shape'[small.df$'cap-shape' == "x"] <- "convex"
small.df$'cap-shape'[small.df$'cap-shape' == "f"] <- "flat"
small.df$'cap-shape'[small.df$'cap-shape' == "k"] <- "knobbed"
small.df$'cap-shape'[small.df$'cap-shape' == "s"] <- "sunken"
# Column 3
levels(small.df$'cap-surface') <- c(levels(small.df$'cap-surface'), "fibrous", "grooves", "scaly", "smooth")
small.df$'cap-surface'[small.df$'cap-surface' == "f"] <- "fibrous"
small.df$'cap-surface'[small.df$'cap-surface' == "g"] <- "grooves"
small.df$'cap-surface'[small.df$'cap-surface' == "y"] <- "scaly"
small.df$'cap-surface'[small.df$'cap-surface' == "s"] <- "smooth"
# Column 6
levels(small.df$'odor') <- c(levels(small.df$'odor'), "almond", "anise", "creosote", "fishy", "foul", "musty", "none", "pungent", "spicy")
small.df$'odor'[small.df$'odor' == "a"] <- "almond"
small.df$'odor'[small.df$'odor' == "l"] <- "anise"
small.df$'odor'[small.df$'odor' == "c"] <- "creosote"
small.df$'odor'[small.df$'odor' == "y"] <- "fishy"
small.df$'odor'[small.df$'odor' == "f"] <- "foul"
small.df$'odor'[small.df$'odor' == "m"] <- "musty"
small.df$'odor'[small.df$'odor' == "n"] <- "none"
small.df$'odor'[small.df$'odor' == "p"] <- "pungent"
small.df$'odor'[small.df$'odor' == "s"] <- "spicy"
tail(small.df)## edible-poison cap-shape cap-surface bruises odor
## 8119 poison knobbed scaly f foul
## 8120 edible knobbed smooth f none
## 8121 edible convex smooth f none
## 8122 edible flat smooth f none
## 8123 poison knobbed scaly f fishy
## 8124 edible convex smooth f noneThis leaves column 5 bruises , which is a True or False field. I could leave t and f, which are easy to understand, spell out true and false, or try to make them logical values. Here is how I made these fields logical.
test <- small.df[[4]] == 't'
small.df[[4]] <- test
tail(small.df)## edible-poison cap-shape cap-surface bruises odor
## 8119 poison knobbed scaly FALSE foul
## 8120 edible knobbed smooth FALSE none
## 8121 edible convex smooth FALSE none
## 8122 edible flat smooth FALSE none
## 8123 poison knobbed scaly FALSE fishy
## 8124 edible convex smooth FALSE noneThis concludes my data transformation of the Mushroom Dataset for this assignment. Here is a summary look of my new dataframe.
summary(small.df)## edible-poison cap-shape cap-surface bruises
## e : 0 convex :3656 scaly :3244 Mode :logical
## p : 0 flat :3152 smooth :2556 FALSE:4748
## edible:4208 knobbed: 828 fibrous:2320 TRUE :3376
## poison:3916 bell : 452 grooves: 4 NA's :0
## sunken : 32 f : 0
## conical: 4 g : 0
## (Other): 0 (Other): 0
## odor
## none :3528
## foul :2160
## fishy : 576
## spicy : 576
## almond : 400
## anise : 400
## (Other): 484