sathi-satb/crime-against-women

Join the “Zero to Data Analyst” Bootcamp starting on March 15. Only 100 seats.

A 20-week program covering 7 courses, 12 assignments, 4 real-world projects, and 12 months of career support.

PS - You need to have the CSV file uploaded in the jupyter notebook. Here is the link to the dataset - https://www.kaggle.com/greeshmagirish/crime-against-women-20012014-india

Crimes against women in India

This project contains the data record of all types of crime against women in india in the span of year 2001-2014. Main aim of working on this project is to reflect the situation of women in our society and raise concern about this matter.

This project is part of my Data Analysis with Python: Zero to Pandas course.

In [31]:

import jovian

In [ ]:

jovian.commit(project='crime-against-women', environment=None)

[jovian] Attempting to save notebook..

In [ ]:

!pip install pandas 
!pip install matplotlib 
!pip install seaborn
!pip install plotly

In [ ]:

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
import plotly.express as px
%matplotlib inline

Importing Data

In [ ]:

jovian.commit('crimes_against_women.csv')

In [ ]:

crimes_df = pd.read_csv('crimes_against_women.csv')

In [ ]:

crimes_df

Let us find out the number of rows and column of the particular dataset

In [ ]:

crimes_df.shape

In [ ]:

import jovian

In [ ]:

jovian.commit()

Data Preparation and Cleaning

To start with the very basic of data cleaning, let's find out if any of the columns have any Null or missing values

In [ ]:

overall_crime = crimes_df.isna().sum()

In [ ]:

overall_crime

None of the column has any Null values.

Now let's find the total number of 'Unique' districts, where the crimes have been committed

In [ ]:

districts = len(crimes_df.DISTRICT.unique())

In [ ]:

districts

But there are 718 districts in India, in total, which means there is messy or false datas in a huge amount, in this case, we better drop the column "District" and also "Unnamed: 0", as it is of no use, in our data analysis process.

In [ ]:

crimes_df.drop(['DISTRICT', 'Unnamed: 0'], axis = 1, inplace=True)

In [ ]:

crimes_df

Now, let's start with analysing the datas of the column "STATE/UT", for that let's find out the names of all the states/UT through .unique()

In [ ]:

print(crimes_df['STATE/UT'].unique())

We can see from above that there are lot many repeated datas, like some of them are repeated again by using capital letters and some of them have issues with space too, like A&N Islands and also Delhi has been repeated again by mentioning it as Delhi UT

In [19]:

# Fist we will remove all the repeated uppercase values
def remove_uppercase(r):
    r = r['STATE/UT'].strip()
    r = r.upper()
    return r
crimes_df['STATE/UT'] = crimes_df.apply(remove_uppercase, axis=1)

#Now use replace function to replace the other type of repeated datas as dicussed above
crimes_df['STATE/UT'].replace("A&N ISLANDS", "A & N ISLANDS", inplace = True)
crimes_df['STATE/UT'].replace("D&N HAVELI", "D & N HAVELI", inplace = True)
crimes_df['STATE/UT'].replace("DELHI UT", "DELHI", inplace = True)

Let's go through the datas now!

In [20]:

crimes_df['STATE/UT'].unique()

Out[20]:

array(['ANDHRA PRADESH', 'ARUNACHAL PRADESH', 'ASSAM', 'BIHAR',
       'CHHATTISGARH', 'GOA', 'GUJARAT', 'HARYANA', 'HIMACHAL PRADESH',
       'JAMMU & KASHMIR', 'JHARKHAND', 'KARNATAKA', 'KERALA',
       'MADHYA PRADESH', 'MAHARASHTRA', 'MANIPUR', 'MEGHALAYA', 'MIZORAM',
       'NAGALAND', 'ODISHA', 'PUNJAB', 'RAJASTHAN', 'SIKKIM',
       'TAMIL NADU', 'TRIPURA', 'UTTAR PRADESH', 'UTTARAKHAND',
       'WEST BENGAL', 'A & N ISLANDS', 'CHANDIGARH', 'D & N HAVELI',
       'DAMAN & DIU', 'DELHI', 'LAKSHADWEEP', 'PUDUCHERRY', 'TELANGANA'],
      dtype=object)

Let's check the total number of States+UT

In [21]:

len(crimes_df['STATE/UT'].unique())

Out[21]:

Which is coming out perfect, hence we are done with our data cleaning process of our dataset

In [22]:

import jovian

In [23]:

jovian.commit()

[jovian] Attempting to save notebook..
[jovian] Updating notebook "sathi-satb/crime-against-women" on https://jovian.ml/
[jovian] Uploading notebook..
[jovian] Capturing environment..
[jovian] Committed successfully! https://jovian.ml/sathi-satb/crime-against-women

Out[23]:

'https://jovian.ml/sathi-satb/crime-against-women'

Exploratory Analysis and Visualization

Let us find out the total population of women over the years, 2001-2014, who has been a victim of the crime based on their gender.

In [24]:

victims_raped = crimes_df.Rape.sum()
victims_kidnapped_abducted = crimes_df.Kidnapping_Abduction.sum()
dowery_death = crimes_df.Dowry_Deaths.sum()
modesty_assault = crimes_df.Assault_for_her_modesty.sum()
insult_to_modesty = crimes_df.Insult_to_modesty_of_Women.sum()
domestic_violence = crimes_df.Domestic_violence.sum()
girls_imported = crimes_df.Importation_of_Girls.sum()

In [25]:

total_population_of_victim_overall = victims_raped + victims_raped + dowery_death +modesty_assault+ insult_to_modesty + domestic_violence+ girls_imported
total_population_of_victim_overall

Out[25]:

This analysis potrays a heartbreaking situation of women in our society, as more than 5 million number of females, over the years 2001-2014, have been a victim of assault, violance, rape or even death, in India alone.

In [26]:

cases_2001_df = crimes_df[crimes_df.Year == 2001]

In [27]:

plt.figure(figsize=(14,6))
plt.hist(cases_2001_df,bins=np.arange(6, 7, 0.25));

In [ ]:

Now let us analyse the all the cases sepately by using bar graph!

In [28]:

fig, axes = plt.subplots(2, 3, figsize=(25, 12))

axes[0,0].set_title("Chart of rape cases in India in 2001-2014")
axes[0,0].bar(crimes_df.Year, crimes_df.Rape, color = 'black');
plt.xlabel('Year') #X-axis
plt.ylabel('Cases of Rape in India') #Y-axis

axes[0,1].set_title("Chart of Kidnapping and Abduction cases in India in 2001-2014")
axes[0,1].bar(crimes_df.Year, crimes_df.Kidnapping_Abduction, color = 'violet');
plt.xlabel('Year') #X-axis
plt.ylabel('Cases of Kidnappinga and Abduction in India') #Y-axis

axes[0,2].set_title("Chart of Dowry death cases in India in 2001-2014")
axes[0,2].bar(crimes_df.Year, crimes_df.Dowry_Deaths, color = 'navy');
plt.xlabel('Year') #X-axis
plt.ylabel('Cases of Dowry deaths in India') #Y-axis

axes[1,0].set_title("Chart of Assault to her modesty in 2001-2014")
axes[1,0].bar(crimes_df.Year, crimes_df.Assault_for_her_modesty, color = 'cyan');
plt.xlabel('Year') #X-axis
plt.ylabel('Cases of Assaulting a women for her modesty in India') #Y-axis

axes[1,1].set_title("Chart of Domestic Violence cases in India in 2001-2014")
axes[1,1].bar(crimes_df.Year, crimes_df.Domestic_violence, color = 'orange');
plt.xlabel('Year') #X-axis
plt.ylabel('Cases of Domestic Violance in India') #Y-axis

axes[1,2].set_title("Chart of Importation of girls in India in 2001-2014")
axes[1,2].bar(crimes_df.Year, crimes_df.Domestic_violence, color = 'red');
plt.xlabel('Year') #X-axis
plt.ylabel('Cases ofImportation of girls in India') #Y-axis

Out[28]:

Text(0, 0.5, 'Cases ofImportation of girls in India')

There are two things to be concluded from the above bar chart -

1) The cases have incraesed over the years.

2) 2014 has been the year, where violance against women was reported the maximum, under each of the cases like, rape, domestic violance etc.

In [29]:

rapecase_data = crimes_df["Rape"]
kidnapping_abd_case = crimes_df["Kidnapping_Abduction"]

In [30]:

plt.pie(crimes_df.Rape, crimes_df.Kidnapping_Abduction, shadow=True, startangle=140)
plt.title("Gold medal achievements of five most successful\n"+"countries in 2016 Summer Olympics")
plt.show()

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
/srv/conda/envs/notebook/lib/python3.7/site-packages/IPython/core/formatters.py in __call__(self, obj)
    339                 pass
    340             else:
--> 341                 return printer(obj)
    342             # Finally look for special method names
    343             method = get_real_method(obj, self.print_method)

/srv/conda/envs/notebook/lib/python3.7/site-packages/IPython/core/pylabtools.py in <lambda>(fig)
    246 
    247     if 'png' in formats:
--> 248         png_formatter.for_type(Figure, lambda fig: print_figure(fig, 'png', **kwargs))
    249     if 'retina' in formats or 'png2x' in formats:
    250         png_formatter.for_type(Figure, lambda fig: retina_figure(fig, **kwargs))

/srv/conda/envs/notebook/lib/python3.7/site-packages/IPython/core/pylabtools.py in print_figure(fig, fmt, bbox_inches, **kwargs)
    130         FigureCanvasBase(fig)
    131 
--> 132     fig.canvas.print_figure(bytes_io, **kw)
    133     data = bytes_io.getvalue()
    134     if fmt == 'svg':

/srv/conda/envs/notebook/lib/python3.7/site-packages/matplotlib/backend_bases.py in print_figure(self, filename, dpi, facecolor, edgecolor, orientation, format, bbox_inches, pad_inches, bbox_extra_artists, backend, **kwargs)
   2215                     orientation=orientation,
   2216                     bbox_inches_restore=_bbox_inches_restore,
-> 2217                     **kwargs)
   2218             finally:
   2219                 if bbox_inches and restore_bbox:

/srv/conda/envs/notebook/lib/python3.7/site-packages/matplotlib/backend_bases.py in wrapper(*args, **kwargs)
   1637             kwargs.pop(arg)
   1638 
-> 1639         return func(*args, **kwargs)
   1640 
   1641     return wrapper

/srv/conda/envs/notebook/lib/python3.7/site-packages/matplotlib/backends/backend_agg.py in print_png(self, filename_or_obj, metadata, pil_kwargs, *args)
    507             *metadata*, including the default 'Software' key.
    508         """
--> 509         FigureCanvasAgg.draw(self)
    510         mpl.image.imsave(
    511             filename_or_obj, self.buffer_rgba(), format="png", origin="upper",

/srv/conda/envs/notebook/lib/python3.7/site-packages/matplotlib/backends/backend_agg.py in draw(self)
    400     def draw(self):
    401         # docstring inherited
--> 402         self.renderer = self.get_renderer(cleared=True)
    403         # Acquire a lock on the shared font cache.
    404         with RendererAgg.lock, \

/srv/conda/envs/notebook/lib/python3.7/site-packages/matplotlib/backends/backend_agg.py in get_renderer(self, cleared)
    416                           and getattr(self, "_lastKey", None) == key)
    417         if not reuse_renderer:
--> 418             self.renderer = RendererAgg(w, h, self.figure.dpi)
    419             self._lastKey = key
    420         elif cleared:

/srv/conda/envs/notebook/lib/python3.7/site-packages/matplotlib/backends/backend_agg.py in __init__(self, width, height, dpi)
     94         self.width = width
     95         self.height = height
---> 96         self._renderer = _RendererAgg(int(width), int(height), dpi)
     97         self._filter_renderers = []
     98 

ValueError: Image size of 1240620x1157162 pixels is too large. It must be less than 2^16 in each direction.

<Figure size 432x288 with 1 Axes>

In [ ]:

In [63]:

import jovian

In [64]:

jovian.commit()

[jovian] Attempting to save notebook..
[jovian] Updating notebook "sathi-satb/crime-against-women" on https://jovian.ml/
[jovian] Uploading notebook..
[jovian] Capturing environment..
[jovian] Committed successfully! https://jovian.ml/sathi-satb/crime-against-women

Out[64]:

'https://jovian.ml/sathi-satb/crime-against-women'

Asking and Answering Questions

In [29]:

import jovian

In [30]:

jovian.commit()

[jovian] Attempting to save notebook..
[jovian] Updating notebook "sathi-satb/crime-against-women" on https://jovian.ml/
[jovian] Uploading notebook..
[jovian] Capturing environment..
[jovian] Committed successfully! https://jovian.ml/sathi-satb/crime-against-women

Out[30]:

'https://jovian.ml/sathi-satb/crime-against-women'

Inferences and Conclusion

In [45]:

import jovian

In [ ]:

jovian.commit()

[jovian] Attempting to save notebook..

References and Future Work

In [ ]:

import jovian

In [ ]:

jovian.commit()

In [ ]: