RUSTAMJI INSTITUTE OF TECHNOLOGY
BSF ACADEMY, TEKANPUR
Synopsis of Minor Project (CS 506)
PREDICTIVE ANALYTICS ON HEALTH CARE
Submitted By
MUSKAN DUSEJA ANUPRIYA SINGH
(0902CS171034) (0902IT171013)
B.Tech Computer Science and Engineering 5th Semester
(2017-2021) Batch
Guided By-
Prof. YOGRAJ SHARMA
Head of Department Computer Science and Engineering
INDEX
Contents Page No.
Team Details 3
Introduction 4
Problem statement 5
Solution statement 6
Technical Details 7-8
Conclusion 9
Bibliography 10
2
Team Details
Muskan Duseja
0902CS171034
Duseja51@gmail.com
Anupriya Singh
0902IT171013
Anupriyasingh317@gmail.com
3
Introduction
This project is based on making predictive analysis based on the data set
using machine learning algorithms. It basically provides you a result on the
basis of the dataset provided by the user. It uses linear regression.
Predictive analytics is the process of learning from historical data in order to
make predictions about the future For health care, predictive analytics will
enable the best decisions to be made, allowing for care to be personalized to
each individual.
We worked on various libraries like numpy , pandas , matplotlib in python.
Then we create our self made library and algorithm for linear regression which
will further do the prediction. The library was then imported and further used
for the process. We worked on Bivariate classification to predict the cases.
The goal of this project is to help doctors make data-driven decisions within
seconds and improve patients' treatment. This is particularly useful in case of
patients with complex medical histories, suffering from multiple conditions.
4
Problem Statement
In healthcare system it is important to predict and analyse healthcare data with faster rates.
big data healthcare analytics is one such approach which address these concerns and issues .
One of the big challenge in present healthcare systems is to take different opinions from
various medical experts as every time it needs a lot of time to understand the previous
patient history and other specific details.
And in today’s world with such a increase in population we might have to wait a long for a
one’s particulars diagnosis treatment.
In simple words , if you are taking a medical treatment and undergoing various medical tests
, clinical reports prescriptions , doctors would take time to understand the yours medical
history and you may be a victim of human error workings and may also face increase in
costs for various treatments process.
And our aim is to reduce costs for treatments by doing prediction on your medical history to
get results in faster rates in order to estimate the varying health condition in emergency
cases .
5
Solution Statement
The solution for the problem statement is given by the prediction done on the basis of
analysis
User had to input dataset on which the software performs algorithms. The
prediction output page will display the medical parameters. We can also compare
several parameter of a individual with a graphical comparison.
It will help the doctors to get immediate results and based on that results the
treatement can be provided to the patients. It will actually save time and
immediate results are shown.
6
Technical Details
Functionality
1. Just a rather simple made gui where you can simply enter the text.
2.After clicking on submit, It will provide you with a result which will be very helpful
to predict the kind of disease that can be caused through it and what kind of
precautions can be taken.
Deployment Environment
Any windows based Operating system can run this program on a desktop or a laptop.
Network Requirements
Internet connection is required to run this program.
Technologies Used
Python (Version 3.0)
Django framework
HTML
7
Third Party Components
BBC articles Dataset
Designing Approach
MACHINE LEARNING
LINEAR REGRESSION FOR BIVARIATE CLASSIFICATION
Screenshots
8
9
Conclusion
We have a that we can use to do prediction on various parameters of
health related issues. This project can be further enhanced by using newer
models.
Its accuracy can be further increased by giving a correct absolute data set .
10
Bibliography
All reference data is taken from
1.Wikipedia
2.Google
3.Python.org
11
12
