Suicide Case Study Worldwide Using Python

By Subhojit Jalal

Here we see a Suicide Case Study, analyze Suicide Cases Worldwide and we calculate the mean value of 100k per Population

Step:-1

First, we import the necessary libraries after that, we upload the dataset

import pandas as pd
import numpy as np
import seaborn as sns
import matplotlib.pyplot as plt
from scipy import stats

Suicide_Dataset = pd.read_csv("Suicide_Dataset.csv")
Suicide_Dataset

Step:-2

Now we plot a graph of who commits Suicide between male and female

Suicide_Dataset_M = Suicide_Dataset[Suicide_Dataset.sex == "male"]
Suicide_Dataset_F = Suicide_Dataset[Suicide_Dataset.sex == "female"]
sns.lineplot(Suicide_Dataset_M.year, Suicide_Dataset.suicides_no, ci = None)
sns.lineplot(Suicide_Dataset_F.year, Suicide_Dataset.suicides_no, ci = None)
plt.legend[("male","female")]
plt.show()

Here we see that female Suicide Case are less than male Suicide Case

Step:-3

Suicide_Dataset_suino = Suicide_Dataset.groupby(["country","year"])["suicides_no"].sum()
Suicide_Dataset_sum = Suicide_Dataset_suino.sort_index(ascending=True)[:] * 100

Suicide_Dataset_pop = Suicide_Dataset.groupby(["country","year"]).population.sum()
Suicide_Dataset_pop_sum = Suicide_Dataset_pop.sort_index(ascending=False)[:]

Suicide_Dataset_total = Suicide_Dataset_sum / Suicide_Dataset_pop_sum
Suicide_Dataset_total.head(10)

Step:-4

Here we see which of countries people are most commit a Suicide 

Country_Dict={}
for country in Suicide_Dataset_total.index.get_level_values(0):
    if country not in Country_Dict.keys():
        Country_Dict[country] = Suicide_Dataset_total[country].mean()
    else:
        pass

pup = list(Country_Dict.items())
pup.sort(key= lambda pair:pair[1], reverse = True)

country_list = [a[0] for a in pup]
country_suicide = [a[1] for a in pup]

plt.figure(figsize=(10,40))
sns.barplot(x=country_suicide[:],y=country_list[:], palette="bright")
plt.title("Rate vs Country")
plt.show()

Step:-5

Now we calculate the mean value of 100k per Population Worldwide also plot a graph of a mean value

Suicide_mean= Suicide_Dataset['suicides/100k pop'].mean()
Suicide_mean

Suicide_mean_data=[]
year=[]
for i in range(len(Suicide_Dataset['year'].unique())-1):
    year.append(Suicide_Dataset['year'].unique()[i])
    Suicide_mean_data.append(Suicide_Dataset[Suicide_Dataset['year']==Suicide_Dataset['year'].unique()[i]]['suicides/100k pop'].mean())

fig = plt.figure(figsize=(15,5))
sns.pointplot(x=year, y=Suicide_mean_data)
plt.axhline(y=Suicide_mean,linestyle='--')
plt.text(2, 13, 'Suicide mean=12.81', fontsize=9, va='center', ha='center')
plt.ylabel("Survivied per 100K people")
plt.xlabel("Year")
plt.title("Worldwide Suicides by Year")

Here we see Suicide at most in 1995

Step:-6

Suicide_sex= Suicide_Dataset[['sex', 'suicides/100k pop']].groupby(['sex'], as_index=False).mean()
Suicide_sex

fig = plt.figure(figsize=(10,5))
labels = 'female','male'
Suicide_sex.plot(kind='pie',labels=labels, y = 'suicides/100k pop', autopct='%1.1f%%')
plt.title('Worldwide Suicides by Gender')

Here we see 100k per Population 79% male and 21% female commit Suicide 

Step:-7

Now we see at which age people commit Suicide most

Suicide_age= Suicide_Dataset[['age', 'suicides/100k pop']].groupby(['age'], as_index=False).mean()
Suicide_age

fig1, ax = plt.subplots()
plt.pie(Suicide_age['suicides/100k pop'], labels=Suicide_age['age'], autopct='%1.1f%%')
plt.title('Worldwide suicides by Age')
plt.axis('equal')
plt.show()

Here we see 31.2% of people are 75+ year old who commits a Suicide Worldwide

Step:-8

Now we see at which generation people commit a Suicide most

Suicide_generation= Suicide_Dataset[['generation', 'suicides/100k pop']].groupby(['generation'], as_index=False).mean()
Suicide_generation

fig1, ax1 = plt.subplots()
plt.pie(Suicide_generation['suicides/100k pop'], labels=Suicide_generation['generation'], autopct='%1.1f%%')
plt.title('Worldwide suicides by Generation')
plt.axis('equal')
plt.show()

Here we see 32.5% of people are G.I Generation who commits a Suicide Worldwide