从数据框中删除因子级别

Question

I downloaded the titanic train dataset from kaggle :

My code is :

df = read.csv('titanic.csv', header=TRUE)
df$Pclass = as.factor(df$Pclass)
df$Survived = as.factor(df$Survived)
df = df[,c(2,3,5,6,12)]
df = na.omit(df)
rownames(df) <- 1:nrow(df)
df$Age[df$Age <= 18] = "child"
df$Age[(df$Age > 18) & (df$Age <= 60) & (df$Age != "child")] = "adult"
df$Age[(df$Age != "child") & (df$Age != "adult")] = "senior"
df$Age = as.factor(df$Age)
summary(df)

At this point the result of summary is :

 Survived Pclass      Sex          Age      Embarked
 0:424    1:186   female:261   adult :553    :  2   
 1:290    2:173   male  :453   child :139   C:130   
          3:355                senior: 22   Q: 28   
                                            S:554 

My problem is Embarked variable:

barplot(table(df$Embarked), xlab="Port of Embarkment", ylab="Frequency", main="Histograma de la variable \n Embarked")

barplot output

The levels of Embarked

> levels(df$Embarked)
[1] ""  "C" "Q" "S"

Here is my problem, this first level : "" (empty) , I can't find a way to remove it. I've been testing several ways I found in stackoverflow without being able to solve my problem.

Answer 1

After removing the lines with empty values for Embarked, refactorize:

df <- df[df$Embarked!="",]
df$Embarked <- factor(df$Embarked)
barplot(table(df$Embarked), xlab="Port of Embarkment", 
        ylab="Frequency", main="Histograma de la variable \n Embarked")

Answer 2

Alternatively, you could also use droplevels :

df <- droplevels(df)

New levels of Embarked :

> levels(df$Embarked)
[1] "C" "Q" "S"

The advantage of this approach it will drop all unused levels from a factor. You can also drop all unused levels from factors in a whole data frame.