-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathSeminar2_solutions3.R
192 lines (140 loc) · 9.75 KB
/
Seminar2_solutions3.R
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
##############################################
## Introductory Data Science for Innovation ##
## Seminar 2 ##
## By Frédérique Bone ##
##############################################
# importing the library for working with tidy data
library(tidyverse) #you may need to install this package if this has not already done before
# Getting the dataset for the seminar
install.packages("titanic") # more info here: https://www.rdocumentation.org/packages/titanic/versions/0.1.0
library(titanic) #call the library
Titanic <- as.data.frame(titanic_train) # this is the main dataset we are going to use throughout the exercise.
#------------------------------
# 1. Explore the given dataset
#------------------------------
## (REPLACE __ by the appropriate value) ##
# 1.1 Look at size of dataset on the right, how many rows and columns?
# 891 rows and 12 columns
# 1.2 You can also use the function 'nrow()' and 'ncol()' on the dataset Titanic
nrow(Titanic) # use this function to find the number of rows
ncol(Titanic) # use this function to find the number of columns
# 1.3 Inspect the columns names of the dataset using 'names()'
names(Titanic) # Use this function to find ou the column names of the dataframe
# 1.4 Use 'head()' and 'tail()' on Titanic to check the 5 first /last rows (using n=) of the data
head(Titanic, n=5)
tail(Titanic, n=8)
# 1.5 Check the overall structure of the dataset use the function 'str()'
# and summary statistics for each variables using 'summary()'
str(Titanic)
summary(Titanic)
# 1.6 Identify which variables may be considered as factors
# 1.7 Change these variables to factor
Titanic$Sex <- as.factor(Titanic$Sex)
class(Titanic$Sex) # check whether this column has been transformed to a factor
Titanic$Embarked <- as.factor(Titanic$Embarked)
class(Titanic$Embarked) # check whether this column has been transformed to a factor
# 1.8 Rerun the 'summary()' function and see what has changed for the two ammended variables
summary(Titanic)
#------------------------------
# 2. Reshape the given dataset
#------------------------------
# 2.0 Remove variables that not of high relevance and has many blanks
# Example of select
Titanic_clean <- Titanic %>% # %>% this is the pipe symbol it enable you to do a series of operations on the dataframes iteratively.
select(-Cabin) # Here we selected out the vairable Cabin as it is not very informative
# Using select with '-' means remove this specific column, use without minus, with a list separated by commas
# to tell the computer which columns you want to keep.
# 2.1 removing, renaming and creating new variables
Titanic_clean <- Titanic %>%
select(-Ticket, -Cabin) %>% # Using 'select' remove the column 'Ticket' and Cabin
rename(Class=Pclass, sibling_spouse=SibSp, parents_children=Parch) %>% # Using the 'rename' function change the name of Pclass to Class,
# SibSp (siblings or spouse onboard) to something more explicit,
# Parch (parents or children onboard) to something more explicit
separate(Name, into=c("Surname","Firstname"), # separate the column 'Name' into 'Surname' and 'Firstname', also looking into the column find the right character to use as a separator
sep=", ", remove=FALSE) %>%
mutate(Overall_relatives = sibling_spouse+parents_children) # Using 'mutate' add a new column 'Overall_relatives' which merges siblings and spouses and parents and children
## If you have time - try to separate again firstname to move out the title e.g. Mr Mrs etc. ##
## If you have time - try to separate again firstname to move out the title e.g. Mr Mrs etc. ##
Titanic_clean <- Titanic_clean %>%
separate(Firstname, into=c("prefix_name","name"), # separate the column 'Name' into 'Surname' and 'Firstname', also looking into the column find the right character to use as a separator
sep="\\. ", remove=FALSE, fill = "left", extra = "merge" )
# 2.2 Practice reshaping dataset.
Titanic_reshape <- Titanic_clean %>% # create a new dataset for reshape
pivot_longer(cols=c(Overall_relatives,sibling_spouse,parents_children), # using pivot longer put all the relatives columns into a unique column, using pivot longer
names_to="family_type", # name the new column 'family_type' (don't forget the quotation marks)
values_to = "family_count") %>% # name the new value column 'family_count'
select(PassengerId, family_type, family_count) # keep only the passenger ids and the 'family_type' and 'family_count'
#------------------------------
# 3. Using logic in your code
#------------------------------
# 3.1 Dealing with unexplicit variables in Embarked (town where people embarked)
unique(Titanic_clean$Embarked) # Use unique to check all the possible values of Embarked
value_embarked <- as.vector(unique(Titanic_clean$Embarked)) # Associate all unique values to the vector and then 'print()' it
print(value_embarked)
# 3.2 Check which entries have missing values
embarked_missing <- Titanic_clean %>% # Create a dataframe containing only these missing values using Titanic_clean
filter (Embarked=="") # using filter, select only the obs. where Embarked is of value ""
head(embarked_missing, n=2) # Using head() inspect the two first rows of this dataframe
# Create a new column in Titanic clean where for the value of the new variable to be entered
Titanic_clean <- Titanic_clean %>%
mutate(Entry_port=NA)
# 3.3 Iterate over the dataset to create the new column with the right
for (i in 1:nrow(Titanic_clean)){ # Create a loop iterating over rows 1 to nrow() of the dataframe to
if (Titanic_clean$Embarked[i] == "S"){ #
Titanic_clean$Entry_port[i] <-"Southampton"
} else if(Titanic_clean$Embarked[i] == "C"){ # create an entry for the C = Cherbourg, Q=Queenstown, ""=NA
Titanic_clean$Entry_port[i] <-"Cherbourg"
} else if(Titanic_clean$Embarked[i] == "Q"){
Titanic_clean$Entry_port[i] <- "Queenstown"
} else if(Titanic_clean$Embarked[i] == ""){
Titanic_clean$Entry_port[i] <-NA
}
}
# 3.4 Remove the variable Embarked using select, make the column 'Entry_port' a factor
# and reassign the results back to Titanic_clean
Titanic_clean <- Titanic_clean %>%
select(-Embarked) %>%
mutate(Entry_port = as.factor(Entry_port))
#--------------------------------
# 4. Dealing with missing values
#--------------------------------
# 4.1 Look again at the summary of Titanic clean to identify the
# column which has NA values
summary(Titanic_clean)
# 4.2 Remove the NAs in for all the relevant variables
Titanic_clean <- Titanic_clean %>% # Use Titanic_clean and re-assign it to itself
filter(!is.na(Entry_port)) %>% # use the right function to select the rows which have NAs for a specific variable
filter(!is.na(Age)) # repeat the above step for all the variables that you identified which have missing NAs
#--------------------------------
# 5. Creating summary statistics
#--------------------------------
# 5.1 Create a dataframe with sum, and count for the people who died or survived
Tit_summary_surv <- Titanic_clean %>% # Use Titanic_clean as a baseline
group_by(Class) %>% # Group by Class
summarise(survived=sum(Survived), # using 'sum()' on survived create a summary on how many survived (using summarise)
Nb_people=n()) %>% # using 'n()' count the number of people who were on the boat
mutate(Perc_died = 100*(Nb_people-survived)/Nb_people) # Generate a percentage of those who died
# 5.2 Create a dataframe to check whether there is a difference in percentage of death for sex (copy and alter the code above)
Titanic_summary_surv_gender <- Titanic_clean %>% # Use Titanic_clean as a baseline
group_by(Sex) %>% # Group by Sex
summarise(survived=sum(Survived), # using 'sum()' on survived create a summary on how many survived (using summarise)
Nb_people=n()) %>% # using 'n()' count the number of people who were on the boat
mutate(Perc_died = 100*(Nb_people-survived)/Nb_people)
# 5.3 Create a dataframe to check whether there is a difference in percentage of death for both sex and class
# Copy and alter the code above, use group_by both sex and class
Titanic_summary_surv_gender_class <- Titanic_clean %>% # Use Titanic_clean as a baseline
group_by(Class, Sex) %>% # Group by Sex
summarise(survived=sum(Survived), # using 'sum()' on survived create a summary on how many survived (using summarise)
Nb_people=n()) %>% # using 'n()' count the number of people who were on the boat
mutate(Perc_died = 100*(Nb_people-survived)/Nb_people)
# 5.4 Create a dataframe which looks at summary for age groups on top of class and sex
# Create categorical variables for Child and Adult
# e.g. Child < 18 and Adult > 18
# Create a summary table to check whether survival in children is more likely than in adults
# Create a new summary table with summary statstics for all Class, Sex and Age
Titanic_summary_surv_child <- Titanic_clean %>%
mutate(child_adult = ifelse(Age < 18, "Child", "Adult")) %>% # Use Titanic_clean as a baseline
group_by(Class, Sex, child_adult) %>% # Group by Sex
summarise(survived=sum(Survived), # using 'sum()' on survived create a summary on how many survived (using summarise)
Nb_people=n()) %>% # using 'n()' count the number of people who were on the boat
mutate(Perc_died = 100*(Nb_people-survived)/Nb_people)