Child Mortality Analysis with Python

This assignment focuses on analyzing global child mortality data using Python. Students will work with a dataset containing information on child mortality rates worldwide. Tasks include data cleaning and preparation, creating informative visualizations (graphs) to depict trends in child mortality, and drawing conclusions based on the analyzed data.

Data
wrangling

Table of Contents
Overview.........................................................................................................................................2
TASK 2...........................................................................................................................................2
PART 1........................................................................................................................................2
PART 2......................................................................................................................................22
Reasons of Child Mortality.........................................................................................................23
Solution for child mortality.........................................................................................................24
Implementation............................................................................................................................24
Conclusion....................................................................................................................................24
References.....................................................................................................................................25
1

Overview
In the project, python Jupiter notebook will be used for implementing child mortality
graphs as per given requirements. The data will be examined, inspected and will be provided as
python code with produced graph screenshots. Further, data wrangling process will be done. The
given csv files will be used and the output screenshots will be produced for child mortality with
numerous graphs.
The statistics is taken from the (WHO) provided link.
TASK 2
The task 2 is separated into two parts:
1. Part 1
2. Part 2
PART 1
Form the given csv file, the python code is produced and executed in part 1. The below
screenshot shows the csv file produced in python code for showing the year, country, Neonatal
mortality rate, infant mortality rate, and Under-five mortality rate. By introducing Pandas library
files in Jupiter platform, the python code is produced further. A brief introduction to python is
provided below
Python
Python is a one of the most widely used high level language and it is fast becoming the
preferred language for aspiring data scientists and for good reasons. It is the most important
language to learn and it also provides the rich ecosystem of a programming language and the
depth of good scientific computation libraries. Its major benefit is breadth.
Python is an interpreter that are available in many of the operating system, object based,
high-level programming language with contain dynamic semantics.it is very simple to learn the
syntax emphasis the readability and hence it decreases the overall production cost of the program
2

maintenance (DataCamp Community, 2017). It care about the modules and packages which
boosts the package modularity and the code can be reuse.
Why python?
Most of the user chose Python in excess of other preferences because Python is easy and
simple to get started with and it can able to handles data wrangling tasks in a simple and
straightforward way. The Python community workings to generate a supportive background for
newcomers.
Python libraries
NumPy
NumPy is abbreviated as Numerical Python. It provides useful features for processes on
n-arrays and matrices in Python. The library offers vectorization of scientific operations on the
NumPy array type, which improves the performance and as a result speeds up the program
execution.
Pandas
The Python data manipulation library Pandas which is mainly used for data manipulation;
for those who are just opening out, this strength imply that this python package that can only be
close when preprocessing the information, but much a smaller amount is only true.
Matplotlib
The library is maintained by various platforms and it makes use of different GUI kits for the
representation of resultant visualizations.
Seaborn
Seaborn is mainly concentrated on the picturing of the statistical models; such type of
visualizations contains heat maps, that can be summarize the information but it still represent the
overall distributions. Seaborn is built on Matplotlib and extremely dependent on that library.
3

4

5

6

7

8

9

10

11

12

13

14

15

16

Charts are produced and shown in the below screenshots. It describes the mortality rates for
various countries such as Africa, Americas and Eastern_Mediterranean various years. The
understandings into the dataset trends are examined and are provided. The variations are shown
for numerous years stated in the requirements.
17

18

19

The child mortality rates under 5 per 1000 live births are obviously produced in python code in
Jupiter notebook.
20

Data Wrangling:
It is about getting complex source of information and then turning this data into useful
information. Once we have cleaned and parsed the information then the datasets are usable. We
can use the methods and tools like python to help to examine the information. And this permits
to get data and produce it accessible and actionable (Kazil, 2017). Data wrangling is also termed
as data munging which is the process of changing and mapping the information from one raw
information to another formation with purpose of producing valuable information like analytics.
It involves information aggregation, data visualization, munging, training a statistical model and
potential uses.
Python:
It is the programming language for multipurpose. It is broadly utilized for the data science,
networking, web development and scientific computing (Babu, 2017).
Jupyter notebook:
21

Now I Python notebook is termed as jupyter notebook. Jupyter notebook is an interactive
computational surroundings in that blend mathematics, rich media, code execution, mathematics
and rich text. We utilize this notebook for analysis of information, graph plotting, data
visualizing and finally sharing the information (Babu, 2017).
Using data wrangling with python jupyter notebook, around worldwide, analyze the mortality
level of the children. We relate the programming of python for the investigation of data. First the
information should be read from the actual sources and the data is wrangled to the formation of
usable information. To get the useful output, we should create the approaches with python
jupyter notebook and solve the analysis issues.
PART 2
json files are nested and explored the data in python. Json is imported
The below screenshot displays clearly
Json file
22

Africa - child mortality rates
The advanced level examination of Africa mortality rate at numerous levels are shown below in
the screenshot.
23

Reasons of Child Mortality
The child are affected by the top 5 causes (Who.int, 2017).
1. Diarrhea (which causes 9 % of deaths)
2. Birth complications (it is 11 %of deaths)
3. Newborn infections, malaria, sepsis, measles (which is 13% of deaths)
4. Birth defects (it is 13 % of deaths)
5. Pneumonia (16 % of deaths)
For the initial stage, a death of neonatal causes which gets highest rank in the overall list. The
next stage, preterm birth complication, and diarrhea is the third stage due to these causes many of
the child deaths. The new rates described 9.2 millions of children under 5 are dying in every
years, from over the 12 million in the year of 1990.the most of the children are dying in the
24

improved countries from the causes of preventable for that there are aware and interventions of
cost effective. Sometimes the efforts are increased there would be less hope of averting the more
over 5.4 million of child dying per year, or decreases of two -third, required to reach millennium
development goal (Howard and Health, 2017).
Solution for child mortality
According to the WHO (world health organization), there are 6 solutions to most preventable
causes under the five deaths, this will includes (Huber, 2017).
1. Exclusive and Immediate breastfeeding.
2. Expert attendants for postnatal care, birth, and antenatal.
3. Access to micronutrients and nutrition.
4. Family information of hazard signs in a child’s health
5. Enhanced access to water, hygiene, and cleanliness.
6. Immunizations.
These solutions are between the different interventions world visualization workers in this work
to encourage nurturing and child healthiness.
Implementation
Conclusion
Python Jupiter notebook is used for implementing child mortality graphs as per given
requirements. The dataset is well examined, inspected and is provided as python code in Jupiter
notebook with produced output screenshots. Data wrangling process is done. The given csv files
is utilized and result is produced for child mortality rate all over the world with numerous
graphs.
25

References
Babu, V. (2017). [online] Available at: https://www.linkedin.com/pulse/learn-python-data-
science-from-scratch-vinay-babu [Accessed 23 Sep. 2017].
DataCamp Community. (2017). Python Exploratory Data Analysis Tutorial. [online] Available
at: https://www.datacamp.com/community/tutorials/exploratory-data-analysis-python#gs._J9idqo
[Accessed 23 Sep. 2017].
Howard, B. and Health, J. (2017). Causes of Child Mortality Released by CHERG. [online]
Johns Hopkins Bloomberg School of Public Health. Available at:
https://www.jhsph.edu/departments/international-health/the-globe/archive/summer2012/child-
causes-of-death.html [Accessed 23 Sep. 2017].
Huber, C. (2017). Child mortality: Top causes, best solutions | World Vision. [online] World
Vision. Available at: https://www.worldvision.org/health-news-stories/child-mortality-causes-
solutions [Accessed 23 Sep. 2017].
Kazil, J. (2017). [online] Available at: http://pdf.th7.cn/down/files/1603/Data%20Wrangling
%20with%20Python.pdf [Accessed 23 Sep. 2017].
Programiz.com. (2017). Learn Python (Programming Tutorial for Beginners). [online] Available
at: https://www.programiz.com/python-programming [Accessed 23 Sep. 2017].
Python Tips. (2017). 20 Python libraries you can’t live without. [online] Available at:
https://pythontips.com/2013/07/30/20-python-libraries-you-cant-live-without/ [Accessed 23 Sep.
2017].
Who.int. (2017). WHO | Child mortality. [online] Available at:
http://www.who.int/pmnch/media/press_materials/fs/fs_mdg4_childmortality/en/ [Accessed 23
Sep. 2017].
26