Analisis Data Eksploratif (EDA) - Titanic Dataset
Annisa Resty
2026-10-05
R Markdown
BAB I: PENDAHULUAN
1.1 Latar Belakang
Tragedi maritim RMS Titanic pada tahun 1912 merupakan salah satu dataset historis yang sering digunakan untuk pembelajaran Sains Data dan Analisis Data Eksploratif (EDA). Sebelum melakukan pemodelan inferensial atau machine learning, tahapan pra-pemrosesan data, pembersihan, pengujian asumsi statistik, serta transformasi fitur sangat krusial untuk memastikan kualitas analisis.
1.2 Tujuan Analisis
- Identifikasi struktur data dan penanganan data hilang (missing values).
- Analisis deskriptif dan visualisasi distribusi variabel usia
(
Age). - Pengujian asumsi parametrik berupa Uji Normalitas (Shapiro-Wilk) dan Uji Homogenitas Varians (Levene’s Test).
- Menerapkan integrasi data dan rekayasa fitur (One-Hot Encoding & Binary Encoding).
BAB II: ANALISIS DATA DAN PEMBAHASAN
Berikut adalah keseluruhan proses analisis data yang dijalankan dalam satu alur eksekusi:
# ==============================================================================
# 1. INSTALL DAN LOAD LIBRARY
# ==============================================================================
required_packages <- c("tidyverse", "car", "knitr")
new_packages <- required_packages[!(required_packages %in% installed.packages()[,"Package"])]
if(length(new_packages)) install.packages(new_packages)
library(tidyverse)
library(car)
library(knitr)
# ==============================================================================
# 2. PENYIAPAN & SIMULASI DATASET TITANIC
# ==============================================================================
set.seed(123)
n <- 891
titanic_raw <- tibble(
PassengerId = 1:n,
Survived = sample(c(0, 1), n, replace = TRUE, prob = c(0.61, 0.39)),
Pclass = sample(c(1, 2, 3), n, replace = TRUE, prob = c(0.24, 0.21, 0.55)),
Name = paste("Passenger", 1:n),
Sex = sample(c("male", "female"), n, replace = TRUE, prob = c(0.65, 0.35)),
Age = c(rgamma(714, shape = 4, scale = 7), rep(NA, 177)),
SibSp = sample(0:5, n, replace = TRUE, prob = c(0.68, 0.23, 0.04, 0.02, 0.02, 0.01)),
Parch = sample(0:5, n, replace = TRUE, prob = c(0.76, 0.13, 0.09, 0.01, 0.005, 0.005)),
Ticket = paste0("TICK_", sample(100:999, n, replace = TRUE)),
Fare = c(rgamma(888, shape = 2, scale = 15), rep(512.3292, 3)),
Cabin = sample(c("C123", "E46", "G6", NA), n, replace = TRUE, prob = c(0.1, 0.1, 0.04, 0.76)),
Embarked = sample(c("S", "C", "Q", NA), n, replace = TRUE, prob = c(0.72, 0.18, 0.09, 0.01))
)
# Struktur Awal Dataset
glimpse(titanic_raw)## Rows: 891
## Columns: 12
## $ PassengerId <int> 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17,…
## $ Survived <dbl> 0, 1, 0, 1, 1, 0, 0, 1, 0, 0, 1, 0, 1, 0, 0, 1, 0, 0, 0, 1…
## $ Pclass <dbl> 1, 1, 3, 2, 1, 3, 3, 3, 2, 2, 3, 2, 1, 1, 3, 1, 2, 2, 3, 3…
## $ Name <chr> "Passenger 1", "Passenger 2", "Passenger 3", "Passenger 4"…
## $ Sex <chr> "female", "male", "male", "male", "female", "male", "femal…
## $ Age <dbl> 31.33544, 21.72096, 47.64552, 33.15069, 49.12994, 63.43629…
## $ SibSp <int> 0, 1, 0, 4, 0, 1, 0, 2, 1, 0, 1, 2, 1, 0, 1, 0, 0, 0, 0, 4…
## $ Parch <int> 0, 0, 0, 0, 0, 2, 0, 1, 0, 0, 0, 0, 0, 1, 2, 0, 0, 2, 0, 1…
## $ Ticket <chr> "TICK_370", "TICK_241", "TICK_563", "TICK_607", "TICK_541"…
## $ Fare <dbl> 23.329218, 36.292773, 42.232290, 58.012474, 28.954200, 20.…
## $ Cabin <chr> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, "E46", "C1…
## $ Embarked <chr> "S", "C", "Q", "S", "C", "S", "C", "S", "S", "Q", "C", "S"…# ==============================================================================
# 3. IDENTIFIKASI MISSING VALUES
# ==============================================================================
missing_summary <- colSums(is.na(titanic_raw))
kable(as.data.frame(missing_summary), col.names = c("Jumlah Missing Value"), caption = "Ringkasan Missing Values Awal")| Jumlah Missing Value | |
|---|---|
| PassengerId | 0 |
| Survived | 0 |
| Pclass | 0 |
| Name | 0 |
| Sex | 0 |
| Age | 177 |
| SibSp | 0 |
| Parch | 0 |
| Ticket | 0 |
| Fare | 0 |
| Cabin | 676 |
| Embarked | 7 |
# ==============================================================================
# 4. DATA CLEANING & IMPUTASI MEDIAN
# ==============================================================================
titanic_selected <- titanic_raw %>%
select(PassengerId, Sex, Age, Survived)
median_age <- median(titanic_selected$Age, na.rm = TRUE)
titanic_clean <- titanic_selected %>%
mutate(Age = if_else(is.na(Age), median_age, Age))
missing_clean <- colSums(is.na(titanic_clean))
kable(as.data.frame(missing_clean), col.names = c("Sisa Missing Value"), caption = "Validasi Setelah Imputasi")| Sisa Missing Value | |
|---|---|
| PassengerId | 0 |
| Sex | 0 |
| Age | 0 |
| Survived | 0 |
# ==============================================================================
# 5. STATISTIK DESKRIPTIF VARIABEL AGE
# ==============================================================================
stat_deskriptif <- titanic_clean %>%
summarise(
Jumlah_Data = n(),
Mean = mean(Age),
Median = median(Age),
Std_Deviasi = sd(Age),
Minimum = min(Age),
Maksimum = max(Age)
)
kable(stat_deskriptif, digits = 2, caption = "Statistik Deskriptif Usia")| Jumlah_Data | Mean | Median | Std_Deviasi | Minimum | Maksimum |
|---|---|---|---|---|---|
| 891 | 27.46 | 25.88 | 12.41 | 3.38 | 100.9 |
# ==============================================================================
# 6. VISUALISASI EKSPLORASI DATA
# ==============================================================================
# A. Histogram & Density Plot
ggplot(titanic_clean, aes(x = Age)) +
geom_histogram(aes(y = ..density..), binwidth = 5, fill = "#4E79A7", color = "white", alpha = 0.7) +
geom_density(color = "#E15759", linewidth = 1) +
labs(title = "Distribusi Usia Penumpang Titanic", x = "Usia (Tahun)", y = "Density") +
theme_minimal()
# B. Q-Q Plot
qqnorm(titanic_clean$Age, main = "Q-Q Plot Usia Penumpang", col = "#4E79A7")
qqline(titanic_clean$Age, col = "#E15759", lwd = 2)
# C. Boxplot Usia berdasarkan Jenis Kelamin
ggplot(titanic_clean, aes(x = Sex, y = Age, fill = Sex)) +
geom_boxplot(alpha = 0.7, outlier.color = "red") +
scale_fill_manual(values = c("female" = "#FF9DA7", "male" = "#4E79A7")) +
labs(title = "Boxplot Usia Berdasarkan Jenis Kelamin", x = "Jenis Kelamin", y = "Usia (Tahun)") +
theme_minimal()
# ==============================================================================
# 7. UJI ASUMSI STATISTIK (NORMALITAS & HOMOGENITAS)
# ==============================================================================
# Uji Normalitas Shapiro-Wilk
shapiro.test(titanic_clean$Age)##
## Shapiro-Wilk normality test
##
## data: titanic_clean$Age
## W = 0.91823, p-value < 2.2e-16# Ringkasan Usia per Jenis Kelamin
stat_sex <- titanic_clean %>%
group_by(Sex) %>%
summarise(
Jumlah = n(),
Mean_Age = mean(Age),
Median_Age = median(Age),
Varians = var(Age)
)
kable(stat_sex, digits = 2, caption = "Ringkasan Usia Berdasarkan Jenis Kelamin")| Sex | Jumlah | Mean_Age | Median_Age | Varians |
|---|---|---|---|---|
| female | 304 | 26.82 | 25.88 | 160.15 |
| male | 587 | 27.79 | 25.88 | 150.65 |
# Uji Homogenitas Varians (Levene's Test)
leveneTest(Age ~ factor(Sex), data = titanic_clean)## Levene's Test for Homogeneity of Variance (center = median)
## Df F value Pr(>F)
## group 1 0.0074 0.9315
## 889# ==============================================================================
# 8. INTEGRASI DATA & CATEGORICAL ENCODING
# ==============================================================================
titanic_integrated <- titanic_clean %>%
select(PassengerId, Sex, Age) %>%
left_join(titanic_clean %>% select(PassengerId, Survived), by = "PassengerId")
titanic_final <- titanic_integrated %>%
mutate(
Sex_binary = if_else(Sex == "male", 1, 0),
Sex_male = if_else(Sex == "male", 1, 0),
Sex_female = if_else(Sex == "female", 1, 0)
)
kable(head(titanic_final, 10), caption = "Dataset Final Setelah Transformasi Encoding")| PassengerId | Sex | Age | Survived | Sex_binary | Sex_male | Sex_female |
|---|---|---|---|---|---|---|
| 1 | female | 31.33544 | 0 | 0 | 0 | 1 |
| 2 | male | 21.72096 | 1 | 1 | 1 | 0 |
| 3 | male | 47.64552 | 0 | 1 | 1 | 0 |
| 4 | male | 33.15069 | 1 | 1 | 1 | 0 |
| 5 | female | 49.12994 | 1 | 0 | 0 | 1 |
| 6 | male | 63.43629 | 0 | 1 | 1 | 0 |
| 7 | female | 18.71087 | 0 | 0 | 0 | 1 |
| 8 | female | 30.13362 | 1 | 0 | 0 | 1 |
| 9 | male | 14.50163 | 0 | 1 | 1 | 0 |
| 10 | male | 61.49842 | 0 | 1 | 1 | 0 |