SEMESTER TRAINING REPORT

on

“Crimes On Women Using Python And Power BI”

Submitted in partial fulfillment of

requirements for the award of the degree

Bachelor in Computer Application

To

Sardar Patial University , Mandi

SUBMITTED BY:
Name: Nimeesha
Roll no.: 2022106012020
Semester: 6th
Batch: 2022-25

Under the guidance of

Mrs. Jyoti Mongra
Assistant Professor

Department of Computer Application

Gian Jyoti Institute of Bachelor Education, Rajol

Kangra, Himachal Pradesh -176208

i
 CERTIFICATE

This is to certify that Ms. Nimeesha has partially completed the Semester Training during the
period from 6th Jan 2025 to 20th Feb 2025 in our Organization as a Partial Fulfilment of
Degree of Bachelor in Computer Application.

(Signature of Project Supervisor)

Date: ____________________

ii
iii
iv
 DECLARATION

I hereby declare that the Project Report entitled (“Crimes On Women Using Python And
Power BI”) is an authentic record of my own work as requirements of 6 th Sem academic
during the period from 6th Jan 2025 to 20th Feb 2025 for the award of degree in BCA, Gian
Jyoti Institute of Bachelor Education, Rajol under the guidance of Ms. Manpreet Kaur.

(Signature of student)
(Nimeesha)
(202201206012020)
Date: ____________________

Certified that the above statement made by the student is correct to the best of our knowledge
and belief.

Signatures

Examined by:

1. 2. 3. 4.

Head of Department
(Signature and Seal)

v
 ACKNOWLEDGMENT
I take this opportunity to express my sincere gratitude to the Principal Mrs. Vijeta
Choudhary, Gian Jyoti Institute of Bachelor Education for providing this opportunity to
carry out the present work.
I am highly grateful to HOD sir of Gian Jyoti Institute of Bachelor Education (Rajol). I would
like to express my gratitude to other faculty members of the BCA department, for providing
academic input, guidance & Encouragement throughout the training period. The help
rendered by Mrs. Jyoti Mongra, Supervisor for Experimentation, is greatly acknowledged.
Finally, I express my indebtedness to all who have directly or indirectly contributed to the
successful completion of my semester training.

vi
 ABOUT COMPANY
As a final semester student pursuing a degree in BCA, I had the opportunity to gain valuable
industry experience through a Six-week Semester Training program with ThinkNEXT
Technologies Pvt. Ltd. During my training, I received a certification in Data Analyst, which
provided me with a strong foundation in programming, software development, and statistical
analysis. However, my major project revolved around the Crimes on Women using Python
and Power BI. This project required extensive work on my part, collaborating with my
project partner, to meet the requirements for our End-Semester Project. Our focus was on
creating an analysis report using Power BI to make it easier to understand the visualized form
of data. The project demanded independent effort outside of the Institute to ensure its
successful completion.

ThinkNEXT Technologies Private Limited, Mohali is an ISO 9001:2015 certified company.

ThinkNEXT was founded in 2011 and is approved from Ministry of Corporate Affairs and
registered under Companies Act 1956. ThinkNEXT deals in Digital Marketing, Web
designing, Web development, Campus ERP Software for Universities/Colleges, eLearning
Platform, Chatbots, Mobile Apps Development, Security Systems, Industrial Training,
Internships, Online Courses etc. ThinkNEXT Technologies provides IT solutions using latest
technologies e.g. Smart Card (Contact Type, Contactless), NFC, Biometrics, Barcode, RFID,
SMS. ThinkNEXT has its numerous clients across the globe. Over the years, with its hard
work, dedication, and commitment, ThinkNEXT has been emerged as one of the leading IT
companies in Chandigarh region. ThinkNEXT is Google Partner for Google AdWords, Bing
Ads Accredited, HubSpot Certified, Facebook Blueprint Certified, Microsoft Bing Ads
Accredited Company. Over the years, with its hard work, dedication, and commitment,
ThinkNEXT is becoming famous in the field of Digital Marketing, Web Development, and
Industrial Training. Founder and Director, Er. Munish Mittal, MTech (IT), BTech (CSE),
himself is having vast experience of more than 20 years in software development and digital
marketing. He is also the author of two books.

vii
 ABSTRACT
Crimes against women remain a pervasive and critical issue globally, affecting their safety,
dignity, and fundamental rights. These crimes encompass a wide range of offenses, including
domestic violence, sexual assault, harassment, human trafficking, honor killings, and dowry-
related violence. Despite significant legal and social advancements, women continue to face
systemic discrimination and violence, often exacerbated by cultural norms, patriarchal
structures, and inadequate enforcement of protective laws. This paper examines the various
forms of crimes against women, their underlying causes, and the socio-legal challenges in
addressing them. The study underscores the importance of a multi-faceted approach to ensure
justice, empowerment, and equality for women in all spheres of life. In this scenario Power
BI play the vital role, because Power BI is an extremely powerful tool for visualizing massive
sets of data very easily. It has an easy to use drag and drop interface. You can build beautiful
visualizations easily and in a short amount of time. Power BI supports a wide array of data
sources. Crimes on women analytics with Power BI, we will create dashboards that helps in
identify the story within our data, and we will better understand the crimes on women cases.

The Power BI are the tools which deals with the big data analytics also it generates the output
in visualization technique with data sources and DAX query language through dashboard in
Power BI,
i.e., more understandable, and presentable. Its features include data blending, real-time
reporting, and collaboration of data. The project aims to show on how we can use the Power
BI with analytics data using DAX query language and its performance on presenting the
dashboard.
This project represents the large dataset into visualization form to quickly see the insights of
our country. In the end, this report gives the clear picture of Crimes on women data and the
tools which can help more effectively, accurately, and efficiently.

viii
 TABLE OF CONTENTS

Title
page .............................................................................................................................................
..
Certificate ...................................................................................................................................
..........
Training Joining
Letter ..........................................................................................................................................
...
Training
Certificate ...................................................................................................................................
...........
Declaration .................................................................................................................................
..........
Acknowledgement ......................................................................................................................
.......................
About
Company ....................................................................................................................................
..........
Abstract ......................................................................................................................................
.......
Table of
Contents ......................................................................................................................................
.......
Chapter 1:
Introduction ...............................................................................................................................
...............
Chapter 2: System
Analysis .......................................................................................................................................
.......
Chapter 3: Software
Tools ............................................................................................................................................
..
Chapter 4: Software
Design .........................................................................................................................................
.....
Chapter 5:
Implementation ..........................................................................................................................
.....................

ix
Chapter 6: Data
Visualization ...............................................................................................................................
................
Chapter 7: Dashboard
Reports .......................................................................................................................................
.........
Chapter 8: Conclusion and Future
work ............................................................................................................................................
....
Chapter 9:
Bibliography ...............................................................................................................................
...................

x
 CHAPTER 1 Introduction

1.1. Brief Overview of Work

Crimes against women in India remain a serious and persistent issue. According to the
National Crime Records Bureau (NCRB), there were over 4.45 lakh cases of crimes against
women reported in 2022. The most common types of crimes include: Power BI’s utility in
analysing and visualizing crimes on women data can be gauged by its performance, user-
friendly environment, and speed. Designed to facilitate the creation of visuals and graphics
without the need for programming knowledge, Power BI democratizes data visualization. It
serves as an intuitive medium for users to easily interpret and analyse data, particularly in the
context of big data.

Power BI’s ability to connect with a variety of data sources and its support for data blending
and realtime collaboration sets it apart as a unique tool. This paper introduces Power BI and
outlines the process of using it for the interactive visualization and analysis of crimes on
women data, advocating for its widespread adoption. As a modern data analytics and
visualization tool, Power BI offers flexibility, ease-of-use, and a smooth experience for users,
enhancing the quality of governance policies and decision-making processes.

1.2. Objective
Crimes against women in India remain a serious and persistent issue. According to the
National Crime Records Bureau (NCRB), there were over 4.45 lakh cases of crimes against
women reported in 2022. The most common types of crimes include:
1. Prevention: Addressing the root causes of violence against women, such as gender
inequality and harmful social norms.
2. Protection: Providing support and protection to survivors of violence, including
access to legal aid, medical care, and counseling.
3. Prosecution: Ensuring that perpetrators of violence are brought to justice and
punished appropriately.

pg. 1
1.3. Scope
The scope of addressing crimes against women includes:
1. Legal reforms: Strengthening existing laws and enacting new laws to protect
women from violence.

2. Law enforcement: Improving the capacity of law enforcement agencies to

investigate and prosecute crimes against women.

3. Social awareness: Raising awareness about the issue of violence against women and
challenging the social norms that perpetuate it.

4. Education: Educating people about gender equality and respect for women.

5. Community involvement: Engaging communities in the fight against violence

against women.

1.4. Project Dashboard

Dashboard describe the crimes on women analysis for the Indian data using various
visualisations and slicers for filtering the data accordingly.

1.5. Project Requirements

Developing a Crimes on women Analysis project necessitates a fusion of suitable software
tools and hardware resources to efficiently gather, handle, scrutinize, and present the vast
datasets implicated. Here is an elaborate breakdown of the software and hardware
prerequisites for constructing such a project:

1.5.1 Hardware Platform

The system requires the following hardware:

• Processor: i3 or above
• Processor speed: 2.00GHz
• CPU RAM: 8GB or above
• Hard disk: 1TB or above
• Internet Connection: Yes

pg. 2
 • Display: 1024 * 768, True Type Color-32 Bit
• Mouse: Any Normal Mouse.
• Keyboard: Any window Supported Keyboard.

1.5.2 Software Platform

The system requires the following hardware:

• IDE: Jupyter Notebook is an open-source IDE used for this project.

• Operation System: Windows or any equivalent can be used for the project.
• Language: Python is a versatile programming language with extensive libraries for
data manipulation and analysis.

• Power BI: Power BI is a business analytics service provided by Microsoft. It aims to

provide interactive visualizations and business intelligence capabilities to create their
own reports and dashboards for the end users.
• Libraries: Different Python libraries listed below are used for cleaning, transforming,
structuring data and for creating visualizations during data exploration.
➢ Pandas
➢ matplotlib.pyplot
➢ seaborn

pg. 3
 CHAPTER 2 System Analysis

AIM and objective of this project is: The aim of the project is to create a report to help the
Crimes on women Analytics to maintain the records of a large data of cases, handle cases
details, and can understand how the cases increased. Crimes on women analytics deals with
the maintenance of cases and mortality rate in India . This Crimes on women analytics system
is user friendly. Having high performance and no time consuming.

2.1 Literature Review

Crimes against women are a global issue rooted in gender inequality, with devastating
consequences for individuals and society. Addressing this requires legal reforms, prevention
programs, support services, and challenging harmful norms.

In this project, we delve into the realm of data analysis using Python. Our goal is to explore
CRIMES ON WOMEN.
19 datasets, extract meaningful insights, and visualize patterns. Let’s walk through the
essential steps:
1. Importing Required Packages:
o Python, with its rich ecosystem of libraries, empowers us. We’ll use Pandas and
NumPy for data wrangling, and Seaborn and Matplotlib for visualization.
o Importing these packages sets the stage for our analysis.
2. Gathering Data:
o Quality data is the bedrock of analysis. We’ve collected CRIMES ON WOMEN data
from various sources.
o Our primary dataset includes cumulative confirmed cases per day in each state.
3. Data Wrangling:
o Raw data needs transformation. We’ll clean, reshape, and prepare it for analysis.

4. Exploratory Data Analysis (EDA) and Visualization:

o EDA unveils insights. We’ll:
▪ Examine time series data.
▪ Visualize patterns.

pg. 4
Importance of Crimes on women Analysis

The analysis of CRIMES ON WOMEN data holds critical importance for several reasons:

1) Understanding the Scope: It reveals the prevalence and trends of these crimes,
highlighting the urgency of the issue.
2) Identifying Root Causes: It helps uncover the underlying factors like gender
inequality, harmful norms, and lack of legal protection.
3) Developing Effective Solutions: It informs the design of targeted interventions,
prevention programs, and support services for survivors.
4) Holding Perpetrators Accountable: It strengthens legal frameworks and improves
enforcement to ensure justice and deter future crimes.
5) Promoting Social Change: It raises awareness, challenges harmful attitudes, and
fosters a culture of respect and equality.
6) Protecting Human Rights: It upholds the fundamental rights of women and girls to
live free from violence and discrimination.

2.2 Project Feasibility Study

2.2.1. Technical Feasibility
Technical feasibility study is concerned with specifying equipment and software that will
successfully satisfy the user requirement; the technical needs of the system may vary
considerably. The facility to produce outputs in each time. Our project is an analysis report
which is based on data provided. In this, every dashboard as output is render from the data so
it is necessary that the dashboard should be rendered in time.

2.2.2. Economical Feasibility

Economical feasibility is the measure to determine the cost and benefit of the proposed
system. A project is economical feasible which is under the estimated cost for its
development. These benefits and costs may be tangible or intangible. Crimes on women
Analysis is the cost-effective project in which there is less possibility of intangible cost so
there is no difficulty to determine the cost of the project.

pg. 5
2.2.3. Operational Feasibility
Operation feasibility is used to check whether the project is operationally feasible or not. Our
project is mainly different from the other system because of its Power BI feature. So, the
measure for operational feasibility is something different from other system. Generally, the
operational feasibility is related to organization aspects.

The change determination is as such that early product were either a man or group of men or
the manual analysis but now a day with the advent of Internet technology.

2.3. Project Timeline Chart

Week 1: Project Initiation and Planning

Week 5: Documentation, and Presentation

Week 2: Data Collection

Week 3: Data Preprocessing

Week 4: Exploratory Data Analysis

( EDA )

Week 5: : Visualisation, Visuals

integration and Dashboard creation

Week 6: Documentation, and Presentation

6
 CHAPTER 3
3.1. Python Libraries
3.1.1. NumPy
NumPy, short for Numerical Python, is an open-source Python library that is integral in the
field of scientific computing. It provides a high-performance multidimensional array object
and tools for working with these arrays. NumPy arrays are faster and more compact than
Python lists, offering an array of operations that can be performed on massive quantities of
data efficiently and with ease.

The library is not only the cornerstone of scientific computing in Python but also a
foundational package that supports a wide array of operations. These include complex
mathematical functions like linear algebra, Fourier transforms, and random number
generation, all of which can be applied to arrays for robust data analysis and manipulation.

NumPy’s capabilities make it a critical tool for researchers, data scientists, and engineers who
require high-level mathematical functions to solve various problems. Its array-processing
package is versatile, allowing it to serve as an efficient multi-dimensional container of
generic data, which is essential for handling the vast datasets commonly encountered in data-
driven fields today.

Moreover, NumPy’s compatibility with a range of other libraries, such as Pandas, SciPy, and
Matplotlib, ensures that it remains a staple in the data analysis workflow, providing a
seamless experience from data preprocessing to visualization3. Its significance is further
underscored by its widespread use in academia and industry, solidifying its position as a
fundamental tool in the Python programming ecosystem.

3.1.2. Pandas
Pandas is a highly esteemed Python library widely utilized for data manipulation and
analysis. It provides high-level data structures and a vast array of tools for cleaning,
transforming, and analysing data. With its core data structure, the DataFrame, Pandas enables
users to manage and manipulate structured data with ease. The library simplifies tasks such as
data import from various file formats, missing data handling, data wrangling, and merging
datasets.

pg. 7
One of the key strengths of Pandas is its ability to work with different data types and
seamlessly manage missing values, which is essential for preparing real-world data for
analysis. It also offers powerful time series functionality, making it the go-to tool for financial
and economic data analysis. Moreover, Pandas integrates well with other libraries in the
Python ecosystem, such as NumPy for numerical computations and Matplotlib for plotting,
creating a robust environment for data science tasks.

Pandas’ syntax is intuitive and readable, which makes it accessible to newcomers while still
being powerful enough for advanced users. Its comprehensive documentation and active
community support further contribute to its popularity among data professionals. Whether it
is for academic research, industry applications, or hobbyist projects, Pandas stands out as a
foundational tool in the Python data analysis landscape.

3.1.3. Matplotlib
Matplotlib is a comprehensive library for creating static, animated, and interactive
visualizations in Python. It is a multi-platform data visualization tool that provides the ability
to plot a wide range of graphs and charts with just a few lines of code. Matplotlib is
particularly useful in scientific computing and data analysis for visualizing complex data
patterns and relationships through its robust plotting functions.
The library’s versatility allows for the creation of simple bar charts, histograms, scatter plots,
and more complex plots like contour plots and 3D graphics. One of Matplotlib’s strengths is
its ability to customize every aspect of a figure, enabling users to achieve the precise look
they desire for their plots. This includes adjusting the colours, line styles, axes, fonts, and
layout.

Matplotlib integrates well with other data science libraries such as NumPy and Pandas,
making it a principal component of the data visualization process in Python. It is also the
foundation upon which libraries like Seaborn and Plotly are built, extending its capabilities
further. Whether for exploratory data analysis, scientific research, or developing sophisticated
data-driven reports, Matplotlib remains a reliable and powerful tool in the Python ecosystem.

3.1.4. Seaborn

pg. 8
Seaborn is a powerful data visualization library in Python that builds on Matplotlib and is
tightly integrated with Pandas data structures. It is designed to create attractive and
informative statistical graphics with ease. Seaborn simplifies the process of generating
complex visualizations like heatmaps, time series, and violin plots, which are essential for
exploring and understanding data patterns and relationships.

The library provides a high-level interface for drawing attractive statistical graphics and is
particularly suited for making complex plots more accessible and understandable. Seaborn’s
default styles and colour palettes are designed to enhance the aesthetic appeal and readability
of plots, which is beneficial when presenting data to an audience.

One of the key features of Seaborn is its ability to work seamlessly with DataFrame objects,
allowing for direct plotting of data from these structures. This makes it an invaluable tool for
data analysts who need to visualize data quickly and effectively. Whether you are a seasoned
data scientist or a beginner, Seaborn’s intuitive syntax and rich functionalities make it an
excellent choice for statistical data visualization in Python.

3.2. Jupyter Notebook

Jupyter Notebook is a versatile web-based platform that combines live code, narrative text,
visualizations, and more. It allows users to create interactive documents where they can write
and execute code in over 40 programming languages, including Python, R, and Julia. These
notebooks are widely used for data analysis, scientific research, and machine learning.
JupyterLab, the nextgeneration interface, offers flexibility and customization, while the
classic Jupyter Notebook provides a streamlined experience. Additionally, Voilà transforms
notebooks into standalone web applications for sharing insights and results. It is an
interactive computing environment that has become indispensable for data scientists,
researchers, and developers. At its core, Jupyter allows users to create documents (called
notebooks) that seamlessly blend code, visualizations, and explanatory text. Jupyter supports
over 40 programming languages, with Python being the most popular. Users can write and
execute code cells within the notebook, making it ideal for data analysis, machine learning,
and scientific computing. Unlike traditional code editors, Jupyter notebooks produce rich
output. This includes visualizations (such as plots and charts), LaTeX equations, images, and
even interactive widgets. It’s a powerful tool for communicating complex ideas. Notebooks

pg. 9
allow users to intersperse code with narrative text. This means you can explain your thought
process, provide context, and document your work—all in one place. It’s like having a lab
notebook for your computational experiments. Jupyter notebooks can be easily shared via
email, GitHub, or other platforms. Colleagues can reproduce your analyses, tweak
parameters, and build upon your work. JupyterHub enables multiuser scenarios, making it
suitable for teams. JupyterLab, the next-generation interface, provides a flexible environment
where users can arrange notebooks, code consoles, and other tools in a customizable layout.
It’s a step forward from the classic notebook interface. In summary, Jupyter Notebook
empowers users to explore data, prototype algorithms, and communicate findings—all within
a single, interactive document.

Features:
1. Live Code Execution:
o Jupyter allows you to write and run code interactively. You can execute code cells and see
the output immediately. o It supports over 40 programming languages, including Python, R,
Julia, Ruby, Scala, and more.
2. Rich Output:
o Jupyter notebooks produce rich output, including text, images, plots, and LaTeX
equations.
o You can visualize data, create charts, and display results in various formats.
3. Narrative Text:
o Markdown cells allow you to add narrative text, headers, lists, and hyperlinks to your
notebook.
o Explain your thought process, document your work, and enhance readability.
4. Interactive Widgets:
o Jupyter supports interactive widgets (e.g., sliders, buttons) that allow users to manipulate
parameters and visualize changes dynamically.
o Useful for exploring data or experimenting with models.
5. Web-Based Interface:
o Jupyter Notebook runs in your web browser, making it accessible from anywhere.
o You can create, edit, and share notebooks online.
6. Kernels:

pg. 10
o Kernels are independent processes that execute user code in specific languages (e.g.,
Python, R).

3.3. Power BI
Power BI is a comprehensive business intelligence platform provided by Microsoft, designed
to transform disparate data sources into coherent and visually immersive insights. It’s a
collection of software services, apps, and connectors that cater to various data analytics
needs, from simple data visualization to complex business intelligence tasks. Power BI
enables users to connect to a multitude of data sources, whether they are simple Excel
spreadsheets or more complex hybrid data warehouses combining cloud-based and on-
premises data.
At its core, Power BI consists of three main components: Power BI Desktop, the Power BI
service, and Power BI Mobile apps. Power BI Desktop is a Windows desktop application that
allows for indepth data analysis and report creation. The Power BI service is an online SaaS
solution that facilitates the sharing and consumption of business insights. Meanwhile, Power
BI Mobile apps extend the platform’s capabilities to iOS, Android, and Windows devices,
ensuring access to data insights on the go.
Features:
Power BI is equipped with a robust set of features that cater to diverse data analytics needs.
Here are some of the key features:
• Monthly Product Updates: Power BI is continuously evolving, with Microsoft adding
new features every month. This ensures that users have access to the latest analytics
capabilities.
• Large Dataset Analysis: Power BI can handle large datasets well beyond the row limits
of Excel, allowing for the analysis of datasets containing over 100 million rows.
• Custom Visualizations: Users can create custom visualizations using R and Python,
which is particularly useful for those who require specialized data representations.
• Excel Integration: Power BI offers deep integration with Excel, enabling users to
analyse their datasets within the familiar environment of Excel spreadsheets (available in
Pro or Premium versions).
• Geospatial Mapping: The platform allows for the creation of interactive maps to display
data geographically, which can be a powerful tool for spatial analysis.

pg. 11
• Power Query: This feature simplifies the process of sourcing and transforming data,
making it easier to prepare data for analysis1.
• Automatic Data Refreshes: Power BI can be set to automatically refresh data, ensuring
that reports and dashboards are always up to date (available in Pro or Premium versions).
• Power BI Mobile App: The mobile app extends the functionality of Power BI to mobile
devices, allowing users to access insights on the go1.
• Dataset Reusability: Users can reuse datasets across different reports and dashboards,
which streamlines the report creation process (available in Pro or Premium versions).
• Microsoft Product Integration: Power BI integrates seamlessly with other Microsoft
products, enhancing its utility within the Microsoft ecosystem.
• These features, among others, make Power BI a versatile and powerful tool for anyone
looking to derive meaningful insights from their data.

pg. 12
 CHAPTER 4 System Design

A system design outline for visualizing crimes against women using Power BI, focusing on
data sources, processing, and visualization:

1. Data Acquisition: Gather crime data from official sources (NCRB, police), NGOs,
surveys, and open data portals.

2. Data Preprocessing: Clean and transform the data, handling missing values,
duplicates, and inconsistencies. Aggregate, calculate metrics, and convert data types.

3. Data Modeling: Design a star schema in Power BI, establishing relationships

between datasets for efficient analysis.

4. Interactive Dashboards: Create dashboards with an overview of key statistics,

drill-down capabilities, and filters for user interaction.

5. Visualizations: Use bar charts, line charts, pie charts, and maps to represent crime
data effectively.

6. Tables and Matrices: Display detailed data and rankings in tables and matrices for
in-depth analysis.

7. KPIs: Track progress and measure the effectiveness of interventions using Key
Performance Indicators.

8. Dashboard Design: Focus on a user-friendly, clear, visually appealing, and

interactive interface.

9. Reporting and Analysis: Generate reports, analyze trends, identify correlations, and
share insights with stakeholders

10. Ethical Considerations: Prioritize data privacy, handle sensitive data responsibly,
and ensure data accuracy.

4.1. System Design Flow Chart

Data Collection

Data pg.
Cleaning
13
 Fig 4.1: System Design Flow Chart

CHAPTER 5 Implementation

pg. 14
The aim of a project on Crimes against Women using Power BI is multifaceted, but boils
down to this core purpose: To provide actionable insights into the prevalence, patterns,
and trends of crimes against women, enabling informed decision-making and targeted
interventions to improve safety and justice.

5.1. Data Source

Retail raw data taken from google.com.
5.1.1 Libraries used

Fig 5.1.1: Libraries Used

5.1.2 Load dataset

pg. 15
 Fig 5.1.2: Displaying file using Pd.read command

5.2 Data Cleaning

Data cleaning in Python is the essential process of preparing raw data for analysis by
handling missing values, inconsistencies, and errors. It's crucial because dirty data leads to
unreliable or misleading results.
5.2.1 Renaming Columns

pg. 16
5.2.2 Dropping unwanted Columns and displaying the clean dataset

5.2.3 Displaying Cleaned Dataset and No. of columns

pg. 17
5.2.4 Displaying the type of data it stores and how much memory is used

5.2.5 Displaying Summary Statistics

pg. 18
5.2.6 Dealing with Null values

5.3 Exploratory data analysis

5.3.1 Plotting Heatmap

pg. 19
5.3.2 Line Chart (or Line Graph) with multiple lines

5.3.3 Bar Chart

pg. 20
5.3.3 Matrix Chart

pg. 21
5.3.4 Small multiples line chart (a panel chart or trellis chart).

5.3.5 Box Plot (also known as a Box and Whisker Plot)

pg. 22
5.4 How to upload raw data into Power BI
To Create Power BI, visualize dashboard, you need to download Power bi desktop and install.
Once you have installed, then we need to follow below process (Microsoft Power BI
documentation, 2021):

Steps for Importing from Files (Excel, CSV, Text, JSON):

1. Open Power BI Desktop: Launch Power BI Desktop.
2. Get Data: On the "Home" ribbon, click "Get Data." A dropdown will appear.
Choose the file type (e.g., Excel, Text/CSV, JSON).
3. Browse: Navigate to the location of your file and select it.
4. Navigator Window: A "Navigator" window will open. Here, you can preview the
data, select which tables or sheets to import (if applicable), and perform basic
transformations (like renaming columns or changing data types) before loading.
5. Load/Transform Data:
 Load: Click "Load" to import the data directly into the Power BI model.
 Transform Data: Click "Transform Data" to open the Power Query Editor.
This is where you can perform more advanced data cleaning, shaping, and
transformation (highly recommended).

pg. 23
Fig 5.4.1: Upload data into Power BI

Fig 5.4.2: Load data into Power BI

pg. 24
 CHAPTER 6 Data Visualization

Power BI is a powerful tool for data visualization, allowing you to create interactive and
insightful reports and dashboards. Here's a breakdown of key aspects:

1. What is Data Visualization in Power BI?

 Interactive visuals: Power BI transforms raw data into engaging visual
representations like charts, graphs, maps, and tables.
 Insightful storytelling: These visuals help you uncover patterns, trends, and
relationships within your data, making it easier to understand and communicate
complex information.
 Data-driven decisions: By presenting data in a clear and concise way, Power BI
empowers you to make informed decisions based on evidence.

2. Types of Visualizations in Power BI

Power BI offers a wide variety of built-in visualizations, including:
 Basic Charts: Bar charts, column charts, line charts, pie charts, area charts
 Advanced Charts: Scatter plots, bubble charts, treemaps, heat maps, waterfall charts
 Maps: Filled maps, bubble maps, shape maps
 Tables and Matrices: Simple tables, pivot tables, matrix visuals
 Other Visuals: Gauges, KPIs, cards, slicers

3. Key Features for Effective Visualization

 Customization: Tailor visuals to your specific needs by adjusting colors, labels, axes,
and formatting.
 Interactivity: Add interactive elements like filters, drill-downs, and tooltips to allow
users to explore data on their own.
 Dashboards: Combine multiple visualizations into a single dashboard to provide a
comprehensive overview of key metrics.
 Data Connectivity: Connect to a wide range of data sources, including Excel files,
databases, and online services.
 AI-powered insights: Leverage AI features to automatically identify trends,
anomalies, and other insights in your data.

4. Best Practices for Data Visualization in Power BI

 Choose the right visual: Select the chart or graph that best suits the type of data you
are presenting and the message you want to convey.
 Keep it simple: Avoid cluttering your visuals with too much information. Focus on
the key insights you want to highlight.
 Use clear labels and titles: Make sure your visuals are easy to understand by using
clear and concise labels and titles.
 Tell a story: Use your visuals to tell a compelling story about your data. Guide your
audience through the insights you have uncovered.
 Test and iterate: Get feedback on your visualizations and make adjustments as
needed to ensure they are clear, effective, and engaging.

pg. 25
6.2 Plotting data:
Data is plotted into several types of graphs in Power BI. According to the need, different
visualisation will be made.

6.2.1. Data Visualisation

• Graph of States of India

In this graph, it shows all the present states of India.

Fig 6.2.1 Slicer (State)

• Graph of cases of Sum of Rape, K&A, DV, DD, AoM

It shows the sum of rape, K&A, DV, DD, AoM cases.

Fig 6.2.2 Cards

pg. 26
 • Graph of cases of DD & DV by Year
In this graph, it shows the sum of DD and DV cases by Year.

Fig 6.2.3 Matrix

• Graph of the Sum of years by State

It shows the State with the total number of cases with the year.

Fig 6.2.4 Donut Chart

• Graph of WT by Year
pg. 27
This graph shows us the sum of WT cases by year.

Fig 6.2.5 Pie Chart

6.3 Creating Dashboards in Power BI

For creating a dashboard in Power BI, various visualisations are combined in a report.
Preparing a dashboard in Power BI involves several key steps to transform your data into a
visual and interactive dashboard. Here’s a high-level overview of the process:

1. Connect to Data Sources: Begin by connecting Power BI to various data sources

such as Excel spreadsheets, databases, or cloud services.
2. Manage Data Model: Organize your data by creating relationships between different
tables and preparing your data model for analysis.
3. Create Visualizations: Select from a range of visualizations like charts, graphs, and
maps to represent your data.
4. Arrange Dashboard: Organize your visualizations logically on the dashboard,
grouping related data together.
5. Enhance Interactivity: Add features like tooltips, drill-downs, and filters to make
your dashboard interactive.

pg. 28
 6. Keep It Simple: Aim for a clean and uncluttered dashboard design to make it easy to
understand.

Initial Dashboard

Fig 6.3: Initial Dashboard

pg. 29
 CHAPTER 7 Dashboard Reports

7.1. Data Dashboard

Fig 7.1: Overall Data Dashboard

7.1.1: Data Dashboard with Bihar’s data

Fig 7.1.2: Data Dashboard

pg. 30
7.1.2: Data Dashboard with Himachal Pradesh’s data

Fig 7.1.2: Data Dashboard

7.1.3: Data Dashboard with Delhi’s data

Fig 7.1.3: Data Dashboard (2)

pg. 31
 CHAPTER 8 Conclusions and Future Scope

8.1. Conclusion
Drawing a definitive conclusion about crimes against women based solely on this single
visualization is impossible. This report provides descriptive statistics, showing the magnitude
of different types of crimes across states and over time. However, it lacks the depth and
context needed for a robust conclusion. Here's what we can say and what further analysis is
required:
What we CAN say based on the visualization:
 High Prevalence: The sheer numbers (hundreds of thousands to millions) indicate a
significant problem of violence against women in India. Each number represents a
victim, highlighting the scale of the issue.
 Varying Crime Types: The report distinguishes several categories of crimes,
suggesting that violence against women takes many forms. This is important for
targeted interventions.
 Geographic Disparities: The state-wise breakdown shows that some regions report
higher numbers of cases than others. This could reflect actual differences in crime
rates, but it could also be influenced by reporting rates, policing practices, or other
factors.
 Potential Temporal Trends: The data spanning from 2006 to 2019 allows for the
observation of potential trends in specific crimes (like Dowry Deaths and Domestic
Violence). However, we can't draw firm conclusions about increasing or decreasing
rates without further statistical analysis.
What we CANNOT say (and what further analysis is needed):
 Root Causes: This report doesn't explain why these crimes occur. Understanding the
underlying social, economic, and cultural factors is crucial for effective prevention.
 Reporting Rates: The numbers likely underestimate the true extent of the problem,
as many cases of violence against women go unreported due to stigma, fear, or lack of
trust in the system. We don't know the reporting rate, so we can't assess the true scope.
 Causality: We cannot conclude from this data that one factor causes another. For
example, we can't say that a particular socioeconomic condition causes higher rates of
domestic violence just by looking at this report.
 Effectiveness of Interventions: The report doesn't provide information on any
programs or policies aimed at addressing violence against women. Therefore, we can't
evaluate their effectiveness.
 Statistical Significance: Observed trends or differences between states could be due
to chance. Statistical tests are needed to determine if the findings are statistically
significant.

pg. 32
 8.2. Benefits
A "Crimes Against Women Report," while dealing with a grim subject, can offer several
crucial benefits when done ethically and responsibly:

1. Awareness and Understanding:

 Public Awareness: These reports bring the issue of violence against women to the
forefront, raising public consciousness and understanding of its prevalence, forms,
and impact.

 Data-Driven Insights: They provide valuable data and statistics that help understand
the scope and nature of the problem, including trends, patterns, and demographics of
victims and perpetrators.

2. Policy and Advocacy:

 Evidence-Based Policy: The data in these reports can inform the development of
effective policies and programs to prevent and address violence against women.

 Resource Allocation: They can guide the allocation of resources to support survivors,
implement prevention programs, and strengthen law enforcement and judicial
systems.

3. Prevention and Intervention:

 Identifying Risk Factors: Analysis of the data can help identify risk factors and
vulnerabilities that contribute to violence against women, enabling targeted
prevention efforts.

 Developing Effective Interventions: Reports can inform the design and

implementation of effective interventions and support services for survivors of
violence.

4. Accountability and Justice:

 Improving Law Enforcement Response: The data can help improve the response of
law enforcement and the judicial system to cases of violence against women.

 Supporting Survivors: By highlighting the experiences of survivors, these reports

can contribute to creating a more supportive environment for them to come forward
and seek help.

5. Research and Knowledge Building:

 Informing Research: Reports provide valuable data for researchers studying violence
against women, contributing to a deeper understanding of the issue.

 Knowledge Sharing: They facilitate knowledge sharing and collaboration among

researchers, practitioners, policymakers, and advocates working to address violence
against women.

pg. 33
8.3. Future Work

The "Crime Against Women Report" you've described, while valuable, can be significantly
enhanced with future work to provide a more comprehensive and actionable understanding of
the issue. Here are some key directions for future development:

1. Deeper Data Disaggregation and Intersectionality:

 Age, Caste, Socioeconomic Background: Break down data not just by state but also
by age group, caste, socioeconomic status, and other relevant demographic factors.
This helps identify the most vulnerable populations and tailor interventions
accordingly.
 Urban vs. Rural: Analyze crime rates separately for urban and rural areas to
understand the unique challenges in each context.
 Intersectionality: Explore how different forms of discrimination intersect and impact
women's experiences of violence (e.g., Dalit women facing both caste-based and
gender-based violence).

2. Enhanced Temporal Analysis:

 Longitudinal Trends: Extend the time series data to identify long-term trends and
patterns in crime rates. This helps assess the effectiveness of existing policies and
programs.
 Seasonal/Cyclical Variations: Investigate if there are seasonal or cyclical variations
in crime rates, which could be linked to festivals, agricultural cycles, or other factors.

3. Contextual Data and Qualitative Insights:

 Socioeconomic Indicators: Incorporate data on poverty, education, employment, and
other socioeconomic indicators to understand the root causes of violence against
women.
 Qualitative Research: Supplement quantitative data with qualitative research, such
as interviews and focus groups, to understand the lived experiences of women facing
violence.
 Law Enforcement Data: Include data on reporting rates, investigation outcomes, and
conviction rates to assess the effectiveness of the criminal justice system.

4. Geographic Information Systems (GIS) Mapping:

 Hotspot Analysis: Use GIS mapping to identify hotspots of crime against women.
This helps target resources and interventions to the areas where they are most needed.
 Spatial Patterns: Analyze spatial patterns of crime to understand the factors that
contribute to violence in specific locations.

pg. 34
5. Predictive Analytics and Modeling:
 Forecasting: Use statistical modeling and machine learning to forecast future trends
in crime against women. This helps policymakers anticipate and prepare for emerging
challenges.
 Risk Factor Identification: Develop models to identify the key risk factors
associated with violence against women. This helps target interventions to the most
vulnerable individuals and communities.

6. Data Visualization and Communication:

 Interactive Dashboards: Develop interactive dashboards that allow users to explore
the data in a user-friendly way.
 Storytelling with Data: Use data visualization techniques to tell compelling stories
about the issue of violence against women. This helps raise awareness and mobilize
action.

7. Collaboration and Data Sharing:

 Inter-agency Collaboration: Foster collaboration between different government
agencies, NGOs, and research institutions to share data and insights.
 Open Data Initiatives: Make the data publicly available through open data initiatives
to encourage further research and analysis.

8. Ethical Considerations:
 Data Privacy and Security: Ensure the privacy and security of data collected from
survivors of violence.
 Avoiding Re-traumatization: Be mindful of the potential for re-traumatization when
collecting and presenting data on violence against women.


By pursuing these directions for future work, the "Crime Against Women Report" can evolve
into a powerful tool for understanding, preventing, and addressing the complex issue of
violence against women in a more effective and meaningful way.

pg. 35
 CHAPTER 9 Bibliography

o https://ncrb.gov.in/

o https://www.unwomen.org/

o https://www.who.int/

o https://www.hrw.org/

o https://www.amnesty.org.in/

o http://www.csrindia.org/

pg. 36