---
title: "Exploratary data analysis for Suicide data set"
output:
flexdashboard::flex_dashboard:
orientation: rows
vertical_layout: scroll
theme: journal
social: menu
source_code: embed
---
```{r setup, include=FALSE}
library(flexdashboard)
library(dplyr)
library(ggplot2)
library(gridExtra)
library(tidyr)
library(reshape2)
library(DT)
data <- read.csv("Suicides_Table_26_2000.csv", stringsAsFactors = FALSE)
#View(data)
# Check for missing values
#colSums(is.na(data))
# Remove rows with missing values
data <- na.omit(data)
#Renaming
names(data) <- gsub(" ", "_", names(data)) #Remove underscores
names(data) <- gsub("[^[:alnum:]_]", "", names(data)) # Remove non-alphanumeric characters
data <- data %>% rename(AlcAddict_M = ByOverAlcoholismMale,
AlcAddict_F = ByOverAlcoholismFemale,
AlcAddict_Total = ByOverAlcoholismTotal,
AlcAddict_Percent=ByOverAlcoholismPercentageshare,
Drowning_M = ByDrowningMale,
Drowning_F = ByDrowningFemale,
Drowning_Total = ByDrowningTotal ,
Drowning_Percent = ByDrowningPercentageshare,
Fire_M =ByFireSelfImmolationMale,
Fire_F = ByFireSelfImmolationFemale,
Fire_Total = ByFireSelfImmolationTotal ,
Fire_Percent = ByFireSelfImmolationPercentageshare ,
Hanging_M = ByHangingMale,
Hanging_F = ByHangingFemale,
Hanging_Total = ByHangingTotal,
Hanging_Percent= ByHangingPercentageshare,
Selfinujury_M=BySelfInflictionofInjuryMale,
SelfInjury_F = BySelfInflictionofInjuryFemale,
SelfInjury_Total = BySelfInflictionofInjuryTotal,
SelfInjury_Percent = BySelfInflictionofInjuryPercentageshare,
Machine_M = ByMachineMale,
Machine_F = ByMachineFemale,
Machine_Total = ByMachineTotal,
Machine_Percent = ByMachinePercentageshare,
SleepingPills_M = ByOverdoseofSleepingPillsMale,
SleepingPills_F=ByOverdoseofSleepingPillsFemale,
SleepingPills_Total = ByOverdoseofSleepingPillsTotal,
SleepingPillls_Percent = ByOverdoseofSleepingPillsPercentageshare,
ElectricWires_M = ByTouchingElectricWiresMale,
ElectricWires_F = ByTouchingElectricWiresFemale,
ElectricWires_Total = ByTouchingElectricWiresTotal,
ElectricWires_Percent=ByTouchingElectricWiresPercentageshare,
TrainFall_M = ByComingunderRunningVehiclesTrainsMale,
TrainFall_F=ByComingunderRunningVehiclesTrainsFemale,
TrainFall_Total=ByComingunderRunningVehiclesTrainsTotal,
TrainFall_Percent=ByComingunderRunningVehiclesTrainsPercentageshare,
Other_M = ByOtherMeansMale,
Other_F = ByOtherMeansFemale,
Other_Total = ByOtherMeansTotal,
Other_Percent = ByOtherMeansPercentageshare,
Total_M=TotalMale,
Total_F=TotalFemale,
Total=Total,
Total_Percent=TotalPercentageshare,
)
# Merge two numeric columns
data$Poison_M <- data$PoisonByConsumingInsecticdesMale + data$ByConsumingOtherPoisonMale
#View(data)
data$Poison_F <- data$PoisonByConsumingInsecticdesFemale + data$ByConsumingOtherPoisonFemale
#View(data)
data$Poison_Total <- data$PoisonByConsumingInsecticdesTotal + data$ByConsumingOtherPoisonTotal
#View(data)
data$Poison_Percent <- data$PoisonByConsumingInsecticdesPercentageshare + data$ByConsumingOtherPoisonPercentageshare
#View(data)
data$SuicidalJump_M <- data$ByJumpingfromBuildingMale + data$ByJumpingfromOtherSitesMale
#View(data)
data$SuicidalJump_F <- data$ByJumpingfromBuildingFemale + data$ByJumpingfromOtherSitesFemale
#View(data)
data$SuicidalJump_Total <- data$ByJumpingfromBuildingTotal + data$ByJumpingfromOtherSitesTotal
#View(data)
data$SuicidalJump_Percent <- data$ByJumpingfromBuildingPercentageshare + data$ByJumpingfromOtherSitesPercentageshare
#View(data)
# Define mergedcolumns to their new positions
columns_to_move <- c("Poison_M", "Poison_F","Poison_Total","Poison_Percent","SuicidalJump_M","SuicidalJump_F","SuicidalJump_Total","SuicidalJump_Percent")
new_positions <- c(26,27,28,29,30,31,32,33)
# Get current column names
cols <- names(data)
# Remove columns to move from their current positions
cols <- setdiff(cols, columns_to_move)
# Insert columns into new positions
for (i in seq_along(columns_to_move)) {
cols <- append(cols, columns_to_move[i], after = new_positions[i] - 1)
}
# Reorder the data frame
data <- data[, cols]
#delete merged cols
cleaned_data <- data %>% select(-PoisonByConsumingInsecticdesMale,-PoisonByConsumingInsecticdesFemale,-PoisonByConsumingInsecticdesTotal,-PoisonByConsumingInsecticdesPercentageshare,-ByConsumingOtherPoisonMale,-ByConsumingOtherPoisonFemale,-ByConsumingOtherPoisonTotal,-ByConsumingOtherPoisonPercentageshare, -ByJumpingfromBuildingMale,-ByJumpingfromBuildingFemale,-ByJumpingfromBuildingTotal,-ByJumpingfromBuildingPercentageshare,-ByJumpingfromOtherSitesMale,-ByJumpingfromOtherSitesFemale,-ByJumpingfromOtherSitesTotal,-ByJumpingfromOtherSitesPercentageshare,-ByJumpingoffMovingVehiclesTrainsMale,-ByJumpingoffMovingVehiclesTrainsFemale,-ByJumpingoffMovingVehiclesTrainsTotal,-ByJumpingoffMovingVehiclesTrainsPercentageshare,-ByFireArmsMale,-ByFireArmsFemale,-ByFireArmsTotal,-ByFireArmsPercentageshare )
```
## Dataset Description {.tabset}
### View of dataset
```{r}
datatable(cleaned_data, extensions = 'Buttons' ,options = list(dom='Bfrtip',Buttons=c('copy','print','PDF','csv')))
```
## Dataset Description {.tabset}
### dataset Structure
```{r}
str(cleaned_data)
```
### Summary
```{r}
summary(cleaned_data)
```
## univariate Analysis {.tabset}
### Histogram-I (Total Suicides)
```{r}
ggplot(cleaned_data, aes(x = Total)) +
geom_histogram(color = "black", fill = "lightblue", binwidth = 100) +
labs(title = "Histogram of Total Suicides", x = "Total Suicides", y = "Frequency")
```
### Histogram-II (Types of Suicides(1-6))
```{r}
plot_1 <- ggplot(cleaned_data, aes(x = AlcAddict_Total)) +
geom_histogram(color = "black", fill = "lightblue", binwidth = 10) +
labs(title = "Alcohol Addiction Suicides", x = "Alcohol Addiction Suicides", y = "Frequency")
plot_2 <- ggplot(cleaned_data, aes(x = Drowning_Total)) +
geom_histogram(color = "black", fill = "lightblue", binwidth = 10) +
labs(title = "Drowning Suicides", x = "Drowning Suicides", y = "Frequency")
plot_3 <- ggplot(cleaned_data, aes(x = Fire_Total)) +
geom_histogram(color = "black", fill = "lightblue", binwidth = 10) +
labs(title = "Fire Suicides", x = "Fire Suicides", y = "Frequency")
plot_4 <- ggplot(cleaned_data, aes(x = Hanging_Total)) +
geom_histogram(color = "black", fill = "lightblue", binwidth = 10) +
labs(title = "Hanging Suicides", x = "Hanging Suicides", y = "Frequency")
plot_5 <- ggplot(cleaned_data, aes(x = Poison_Total)) +
geom_histogram(color = "black", fill = "lightblue", binwidth = 10) +
labs(title = "Poison Suicides", x = "Poison Suicides", y = "Frequency")
plot_6 <- ggplot(cleaned_data, aes(x = SelfInjury_Total)) +
geom_histogram(color = "black", fill = "lightblue", binwidth = 10) +
labs(title = "Self-Injury Suicides", x = "Self-Injury Suicides", y = "Frequency")
grid.arrange(plot_1, plot_2, plot_3, plot_4, plot_5, plot_6, ncol = 3)
```
### Histogram-III (Types of Suicides(7-12)
```{r}
plot_7 <- ggplot(cleaned_data, aes(x = Machine_Total)) +
geom_histogram(color = "black", fill = "lightblue", binwidth = 10) +
labs(title = "Machine Suicides", x = "Machine Suicides", y = "Frequency")
plot_8 <- ggplot(cleaned_data, aes(x = SleepingPills_Total)) +
geom_histogram(color = "black", fill = "lightblue", binwidth = 10) +
labs(title = "Sleeping Pills Suicides", x = "Sleeping Pills Suicides", y = "Frequency")
plot_9 <- ggplot(cleaned_data, aes(x = ElectricWires_Total)) +
geom_histogram(color = "black", fill = "lightblue", binwidth = 10) +
labs(title = "Electric Wires Suicides", x = "Electric Wires Suicides", y = "Frequency")
plot_10 <- ggplot(cleaned_data, aes(x = TrainFall_Total)) +
geom_histogram(color = "black", fill = "lightblue", binwidth = 10) +
labs(title = "Train Fall Suicides", x = "Train Fall Suicides", y = "Frequency")
plot_11 <- ggplot(cleaned_data, aes(x = Other_Total)) +
geom_histogram(color = "black", fill = "lightblue", binwidth = 10) +
labs(title = "Other Suicides", x = "Other Suicides", y = "Frequency")
plot_12 <- ggplot(cleaned_data, aes(x = SuicidalJump_Total)) +
geom_histogram(color = "black", fill = "lightblue", binwidth = 10) +
labs(title = "Suicidal Jump Suicides", x = "Suicidal Jump Suicides", y = "Frequency")
grid.arrange(plot_7, plot_8, plot_9, plot_10, plot_11, plot_12, ncol = 3)
```
```{r}
# Subset State, UT, City
Statewise_data <- cleaned_data %>% filter(Category == "State")
UTwise_data <- cleaned_data %>% filter(Category == "UT")
Citywise_data <- cleaned_data %>% filter(Category == "City")
# Subset Top 5 States, Cities, UTs
top_5_states <- Statewise_data %>%
arrange(desc(Total)) %>%
head(5)
top_5_uts <- UTwise_data %>%
arrange(desc(Total)) %>%
head(5)
top_5_cities <- Citywise_data %>%
arrange(desc(Total)) %>%
head(5)
# Subset data by method of suicide
alc_addict <- cleaned_data %>% filter(AlcAddict_Total > 0)
drowning <- cleaned_data %>% filter(Drowning_Total > 0)
fire <- cleaned_data %>% filter(Fire_Total > 0)
hanging <- cleaned_data %>% filter(Hanging_Total > 0)
poison <- cleaned_data %>% filter(Poison_Total > 0)
self_injury <- cleaned_data %>% filter(SelfInjury_Total > 0)
machine <- cleaned_data %>% filter(Machine_Total > 0)
sleeping_pills <- cleaned_data %>% filter(SleepingPills_Total > 0)
electric_wires <- cleaned_data %>% filter(ElectricWires_Total > 0)
train_fall <- cleaned_data %>% filter(TrainFall_Total > 0)
other <- cleaned_data %>% filter(Other_Total > 0)
suicidal_jump <- cleaned_data %>% filter(SuicidalJump_Total > 0)
```
### Suicides(Citywise)
```{r}
ggplot(Citywise_data, aes(x = reorder(StateUTCity, Total), y = Total)) +
geom_point(aes(color = Total), size = 3) +
labs(title = "Scatter Plot of Suicides by City", x = "City", y = "Total Suicides") +
theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
scale_color_viridis_c()
```
### Suicides(Statewise)
```{r}
ggplot(Statewise_data, aes(x = reorder(StateUTCity, Total), y = Total)) +
geom_point(aes(color = Total), size = 3) +
labs(title = "Scatter Plot of Suicides by States", x = "State", y = "Total Suicides") +
theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
scale_color_viridis_c()
```
### Suicides(UTwise)
```{r}
ggplot(UTwise_data, aes(x = reorder(StateUTCity, Total), y = Total)) +
geom_point(aes(color = Total), size = 3) +
labs(title = "Scatter Plot of Suicides by UT", x = "UT", y = "Total Suicides") +
theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
scale_color_viridis_c()
```
### Boxplot(Male)
```{r}
ggplot(cleaned_data, aes(x = "", y = Total_M)) +
geom_boxplot(fill = "lightblue") +
labs(title = "Box Plot of Male Suicides", x = "", y = "Number of Male Suicides") +
theme_minimal() +
theme(axis.title.x = element_blank(), axis.text.x = element_blank())
```
### Boxplot(Female)
```{r}
ggplot(cleaned_data, aes(x = "", y = Total_F)) +
geom_boxplot(fill = "lightblue") +
labs(title = "Box Plot of Female Suicides", x = "", y = "Number of Female Suicides") +
theme_minimal() +
theme(axis.title.x = element_blank(), axis.text.x = element_blank())
```
### Types_of_Suicide
```{r}
totals_long <- cleaned_data %>%
select(contains("_Total")) %>%
pivot_longer(
cols = everything(),
names_to = "Method",
values_to = "Total"
) %>%
mutate(Method = gsub("_Total", "", Method)) %>%
mutate(Total = as.numeric(Total)) %>%
select(Method, Total)
totals_long_filtered <- totals_long %>%
filter(Total > 0)
totals_long_filtered <- totals_long_filtered %>%
filter(Total < 10000)
ggplot(totals_long_filtered, aes(x = reorder(Method, -Total), y = Total)) +
geom_bar(stat = "identity", fill = "salmon") +
labs(title = "Total Suicides by Method", x = "Method", y = "Total Suicides") +
theme(axis.text.x = element_text(angle = 90, hjust = 1)) # Rotate x-axis labels for readability
```
