Project ID: SPS_PRO_215 GitHub Title: llSPS-INT-1684 1
Predicting Life Expectancy using
Machine Learning
SB42048
Nishant Saxena
3rd Year Computer Science Engineering NMIMS
PROJECT ID: SPS_PRO_215 GITHUB TITLE: LLSPS-INT-1684 2
Introduction
Overview
Machine Learning project leverage historical data to predict insights into the future. In this
project by the problem statement, we have to predict the Life Expectancy of individuals based
on various features. Life Expectancy is a statistical measure of the average time a human being
is expected to survive. Life expectancy depends on various factors, in this problem we will
predict the Life expectancy of people living in a country when factors like the year, GDP,
Measles, Alcohol, Thinness, etc. are given.
Purpose
Life expectancy plays an important role when decisions about the final phase of life need to be
made. Good prognostication for example helps to determine the course of treatment and helps
to anticipate the procurement of health care services and facilities, or more broadly: facilitates
Advance Care Planning.
Literature Survey
Existing Problem
This problem statement gives an approach to foresee the normal future of individuals living in a
nation when different factors, for example, year, GDP, liquor admission of individuals in the
nation, and some particular infection-related death that occurred in the nation are given.
Life expectancy is a statistical measure of the average time a human being is expected to live,
Life expectancy depends on various factors.
PROJECT ID: SPS_PRO_215 GITHUB TITLE: LLSPS-INT-1684 3
Proposed Solution
Design a Machine Learning Model to predict Life Expectancy using the features given in any
country.
Theoretical Analysis
Block Diagram
Hardware / Software Designing
▪ Hardware: Desktop / Laptop, Internet Connection.
▪ Software: IBM Cloud Account, IBM Watson Studio, Node-Red App.
Literature Survey
1. Collecting Dataset from
https://www.kaggle.com/kumarajarshi/life-expectancy-who
PROJECT ID: SPS_PRO_215 GITHUB TITLE: LLSPS-INT-1684 4
2. Study the data and create a hypothesis on how can we predict the life
expectancy of a person.
3. Predict Life Expectancy using different machine learning models and select the
model that gives results with maximum accuracy.
4. Hence we will be able to predict the life expectancy of a person with maximum
accuracy.
PROJECT ID: SPS_PRO_215 GITHUB TITLE: LLSPS-INT-1684 5
Theoretical Analysis
Block Diagram
PROJECT ID: SPS_PRO_215 GITHUB TITLE: LLSPS-INT-1684 6
PROJECT ID: SPS_PRO_215 GITHUB TITLE: LLSPS-INT-1684 7
Result
PROJECT ID: SPS_PRO_215 GITHUB TITLE: LLSPS-INT-1684 8
Advantages and Disadvantages
Advantages
▪ IBM Watson offers many advance tools to construct Machine learning model
▪ With Node-Red we can easily make UI.
▪ IBM Cloud is easy to use Platform.
▪ Don’t require external storage
Disadvantages
▪ Requires Internet Connection
▪ With the Lite plan, you can only deploy one app at once.
Application
Life Expectancy is the essential factor in deciding a person's risk factor and the probability they
will make a case. Insurance agencies think about age, way of life decisions, and a few different
components while deciding premium rates for singular life coverage strategies. It very well may
be utilized by specialists to make important inquires about out of it and along these lines, realize
something that will help increment the hope think about the effect of a particular factor on the
normal life expectancy of individuals in a particular nation.
Conclusion
Consequently, we have built up a model that will foresee the future of individuals and predict its
Life Expectancy of a specific demographic region based on provided constraints. Different
variables significantly affect life expectancy, for example, Adult Mortality, Population, Under 5
Deaths, Thinness 1-5 Years, Alcohol, HIV, Hepatitis B, GDP, Percentage Expenditure, and
PROJECT ID: SPS_PRO_215 GITHUB TITLE: LLSPS-INT-1684 9
some more. The client can connect with the framework through a straightforward UI which is a
form with input spaces that the client needs to fill the contributions to.
Future Scope
By deploying it we can store the data of individuals filling the form to determine the variation of
values of different variables to improve the model and to make it better.
