Great Deal! Get Instant $10 FREE in Account on First Order + 10% Cashback on Every Order Order Now

Introduction Healthcare is a crucial aspect of modern life, affecting the well-being and longevity of individuals worldwide. Access to affordable and high-quality healthcare is a major concern...

1 answer below »
Introduction
Healthcare is a crucial aspect of modern life, affecting the well-being and longevity of individuals
worldwide. Access to affordable and high-quality healthcare is a major concern for people in every
corner of the globe, and the cost of medical treatment can pose a significant financial burden,
especially for those who are not covered by insurance. While healthcare costs continue to rise,
insurance companies are tasked with finding ways to minimize risk and optimize costs to provide their
clients with the best possible coverage. To achieve this goal, insurers need to have a comprehensive
understanding of their clients' health status, lifestyle habits, and other relevant factors that may
impact their health.

In recent years, the healthcare industry has undergone significant changes and innovations, thanks to
the development of new technologies and the application of data-driven insights. These
advancements have helped healthcare providers and insurance companies better understand the
healthcare needs of their clients, and have enabled them to deliver more targeted and personalized
care. By leveraging data analytics, insurers can gain valuable insights into their clients' health status,
which allows them to develop more effective and efficient insurance plans that meet the needs of
their clients.

One of the key challenges that insurance companies face in the healthcare industry is the need to
minimize risk while still providing comprehensive coverage to their clients. This requires a proactive
approach to healthcare, where insurers work closely with their clients to understand their health
needs and risks. By collecting and analysing data on their clients' health status, lifestyle habits, and
other relevant factors, insurers can identify potential risks and develop strategies to mitigate them.
For example, insurers may offer incentives to encourage healthy behaviour, such as discounts on gym
memberships or wellness programs. By promoting healthy habits, insurers can help reduce the risk of
illness and disease, which ultimately benefits both the insurer and the client.

In addition to promoting healthy habits, insurers must also be proactive in managing chronic
conditions. Chronic conditions, such as diabetes and heart disease, can be costly to treat and can pose
a significant risk to insurers. By providing targeted support and resources to clients with chronic
conditions, insurers can help their clients manage their health more effectively and reduce the overall
cost of care. This may include providing access to specialized care teams, offering disease
management programs, and providing incentives for medication adherence and regular check-ups.

Data analytics is a critical tool for insurers looking to optimize their costs and minimize risk in the
healthcare industry. By collecting and analysing data on their clients' health status and healthcare
utilization, insurers can gain valuable insights into their clients' needs and risks. This data can be used
to develop more effective insurance plans that meet the specific needs of their clients, while also
minimizing risk and reducing costs. For example, insurers can use data analytics to identify high-risk
clients and develop targeted interventions to help manage their health more effectively. This may
include offering personalized health coaching, providing access to specialized care teams, or offering
financial incentives for healthy behaviour.

In addition to using data analytics to optimize costs, insurers can also leverage technology to improve
the overall quality of care. Telemedicine and other virtual care solutions have become increasingly
popular in recent years, as they provide a more convenient and cost-effective way for patients to
access healthcare services. By offering virtual care solutions, insurers can help their clients access care
more easily and cost-effectively, which can ultimately improve their health outcomes.

While technology and data analytics have the potential to transform the healthcare industry, they also
pose significant challenges. Data privacy and security are critical concerns, as the collection and use
of personal health data must be done in a way that is both ethical and secure. Insurers must also
ensure that their virtual care solutions and other technological innovations are accessible to all clients,
egardless of their income or technological literacy.

In conclusion, healthcare is a crucial aspect of modern life, impacting the well-being and longevity of
individuals worldwide. Access to affordable and high-quality healthcare is a major concern

BusinessProblem
We all know that Health care is very important domain in the market. It is directly linked with the life
of the individual; hence we have to be always be proactive in this particular domain. Money plays a
major role in this domain, because sometime treatment becomes super costly and if any individual is
not covered under the insurance then it will become a pretty tough financial situation for that
individual. The companies in the medical insurance also want to reduce their risk by optimizing the
insurance cost, because we all know a healthy body is in the hand of the individual only. If individual
eat healthy and do proper exercise the chance of getting ill is drastically reduced.

Goal&Objective:
The objective of this exercise is to build a model, using data that provide the optimum insurance cost
for an individual. You have to use the health and habit related parameters for the estimated cost of
insurance
File:Data.csv
Targetvariable:insurance_cost

Data Report


Missing Value


Replacing missing values in BMI column from mean and dropping the Applicant ID and year last admit
ecause in year last admit there is lot of missing values which means that patient never admit before
this study so the reason of removal of these column is creating extra dimension to the data frame
while converting it into dummy variable so it is better to remove the remove the column to to avoid
drastic increase in the data dimension.


Now from the above box plot we can see that the variables do not contain much extreme values in
the data So we can perform models on the given data and if it not works then we come to remove
lawyers from the data based on quartile deviation or standard deviation.


From He
ew pair plot we can see that insurance and weight are perfectly positively co
elated while
all other variables not seem like perfect co
elated so we will perform co
elation matrix numerical
variable to check weather independent variable are highly co
elated with each other or not if they
are highly co
elated then we will remove that column

From the above co
elation matrix, we can see that our numerical data is not highly co
elated with
each other So we will so we will continue with the given data for further investigation











From the eBook data description we can see that the age is ranging from 274 add the average cost of
insurance is 27,147 and the maximum will visited doctor in last one year is 12 and the average daily
steps covered by the patient is 5000.
















Converting categorical variable into dummy variable and treating first column as a reference variable


After the conversion of categorical variable into dummy variable now we will separate the target
variable and the dependent variable so that we can perform model on it.


Show the given problem is the regression problem so we will use regression models to predict
insurance costs half a patient covered by insurance companies should there are many different models
we will use some of them are
o linear regression
o rigid regression
o lasso
o elastic net
o polynomial regression
o k nearest neighbour
o support vector machine
o decision tree regression
o random forest regressor








elow is the given result based on the above models








From the table below we can see that R2 score and best parameter of the models and we will select
est model from them

Best Model selection




Random forest is the best performing model among the all model


PowerPoint Presentation
Capstone Presentation
Guidelines to be followed :
Presentation should complete in 10 mins
5 minutes will be devoted to Q&A
Business Problem Understanding
Business problem we are trying to solve
constraints
scope
objectives
Mention whatever is applicable
Tentative time – 2.5 mins
Modelling Approach Used & Why
Tentative time – 2.5 mins
Subheading
Lorem Ipsum is simply dummy text of the printing and typesetting industry.
Insights from Analysis
Tentative time – 2.5 mins
 
Recommendations
Tentative time – 2.5 mins

Go Back to Capstone Project - PGP-DSBA
Course Content
Final PPT submission
Submission type : File Upload
Due Date : Mar 10, 11:59 PM
Total Marks : 20
Available from : Mar 03, 8:00 AM
Description
Dear Learner,
You need to submit the Final Presentation Here.
Do follow the guidelines mention in the module in regards to how to create PPT. You need to
use PPT template.
The Presentation is of 20 marks which will include
1. Presentation Submission
2. Presentation of the deck in front of the mentor as per the slots given.
3. Viva in regards to the Project and methodologies applied
Regards,
Program Office
Scoring guide (Ru
ic) - Some Ru
ic (2) Evaluated
Criteria Ratings Points
1.1 Business Problem
Understanding (Explanation)
4
Final PPT submission: Capstone Project - PGP-DSBA - G... https:
olympus.mygreatlearning.com/courses/74432/as...
1 of 3 3/10/23, 06:45
https:
olympus.mygreatlearning.com/courses/74432
https:
olympus.mygreatlearning.com/courses/74432
https:
olympus.mygreatlearning.com/courses/74432
https:
olympus.mygreatlearning.com/courses/74432
https:
olympus.mygreatlearning.com/courses/74432
https:
olympus.mygreatlearning.com/courses/74432
https:
olympus.mygreatlearning.com
https:
olympus.mygreatlearning.com
Proprietary content.©Great Learning. All Rights Reserved. Unauthorized use or distribution prohibited.
© 2023 All rights reserved Privacy Terms of service Help
Understanding (Explanation)
1.2 Business Problem
Understanding (ppt)
2
2.1 Modeling approach used & why
(Explanation)
5
2.2 Modeling approach used & why
(ppt)
3
3.1 Insights & recommendation
(Explanation)
4
3.2 Insights & recommendations
(ppt)
2
Points null/20
Submit Assignment
Upload file
Submit Assignment
Previous Next
Add comments
Final PPT submission: Capstone Project - PGP-DSBA - G... https:
olympus.mygreatlearning.com/courses/74432/as...
2 of 3 3/10/23, 06:45
https:
www.mygreatlearning.com/privacy-policy
https:
www.mygreatlearning.com/privacy-policy
https:
www.mygreatlearning.com/privacy-policy
https:
www.mygreatlearning.com/terms
https:
www.mygreatlearning.com/terms
https:
www.mygreatlearning.com/terms
https:
olympus1.mygreatlearning.com/program_support?pb_id=12907
https:
olympus1.mygreatlearning.com/program_support?pb_id=12907
https:
olympus1.mygreatlearning.com/program_support?pb_id=12907
https:
olympus.mygreatlearning.com/courses/74432/pages/capstone-project-presentation-2?module_item_id=2343287
https:
olympus.mygreatlearning.com/courses/74432/files/4888135?module_item_id=2343289
https:
olympus.mygreatlearning.com/courses/74432/pages/capstone-project-presentation-2?module_item_id=2343287
https:
olympus.mygreatlearning.com/courses/74432/pages/capstone-project-presentation-2?module_item_id=2343287
https:
olympus.mygreatlearning.com/courses/74432/pages/capstone-project-presentation-2?module_item_id=2343287
https:
olympus.mygreatlearning.com/courses/74432/pages/capstone-project-presentation-2?module_item_id=2343287
https:
olympus.mygreatlearning.com/courses/74432/pages/capstone-project-presentation-2?module_item_id=2343287
https:
olympus.mygreatlearning.com/courses/74432/files/4888135?module_item_id=2343289
https:
olympus.mygreatlearning.com/courses/74432/files/4888135?module_item_id=2343289
https:
olympus.mygreatlearning.com/courses/74432/files/4888135?module_item_id=2343289
https:
olympus.mygreatlearning.com/courses/74432/files/4888135?module_item_id=2343289
https:
olympus.mygreatlearning.com/courses/74432/files/4888135?module_item_id=2343289
Final PPT submission: Capstone Project - PGP-DSBA - G... https:
olympus.mygreatlearning.com/courses/74432/as...
3 of 3 3/10/23, 06:45
https:
olympus.mygreatlearning.com/courses/74432/pages/capstone-project-presentation-2?module_item_id=2343287
https:
olympus.mygreatlearning.com/courses/74432/files/4888135?module_item_id=2343289
https:
olympus.mygreatlearning.com/courses/74432/pages/capstone-project-presentation-2?module_item_id=2343287
https:
olympus.mygreatlearning.com/courses/74432/pages/capstone-project-presentation-2?module_item_id=2343287
https:
olympus.mygreatlearning.com/courses/74432/pages/capstone-project-presentation-2?module_item_id=2343287
https:
olympus.mygreatlearning.com/courses/74432/pages/capstone-project-presentation-2?module_item_id=2343287
https:
olympus.mygreatlearning.com/courses/74432/files/4888135?module_item_id=2343289
https:
olympus.mygreatlearning.com/courses/74432/files/4888135?module_item_id=2343289
https:
olympus.mygreatlearning.com/courses/74432/files/4888135?module_item_id=2343289
https:
olympus.mygreatlearning.com/courses/74432/files/4888135?module_item_id=2343289
Answered 4 days After Mar 10, 2023

Solution

Mukesh answered on Mar 14 2023
32 Votes
Capstone Project – NBFC Loan Foreclosure PN2
Capstone Project – Heealthcare PN2
-    N I M E S H M A R F A T I A
Agenda
Discuss the feedback of the evaluation of last note submission.
Discuss Model Building
Discuss Model Tuning & Validation
Discuss on Model Interpretation
Discuss about how to draw Business Insights & Recommendations.
Assignment Notes 2 Evaluation
Review Parameters
1) Model building and interpretation.
Build various models
Test your predictive model against the test set using various appropriate performance metrics
Interpretation of the models
Review Points
10
2) Model Tuning
Ensemble modelling, wherever applicable
Any other model tuning measures(if applicable)
Interpretation of the most optimum and its implication on business
Total
10
20
Problem Statement, Scope & Business Objective:
Problem Statement:
The given dataset has multiple parameters which influence the health of the person and in turn impacts his / her insurance premium (Cost of Insurance);
The given scenario advises us to identify the parameters and provide weightage for each and determine the optimal insurance premium
for a person covering his/her risk
Goal & Objective –
The objective of this project is to build a model, using the health and habit parameters in the dataset and provide the optimum insurance
cost for an individual.
Optimal & cost effective premiums results in more market Share for the enterprise, more Profits and enhances Branding of the Business.
Limits the wealth erosion, better predictability and improves the standard of living
EDA: Create Data Set for your Model
Missing Values:
Imputation of more than 10-15% is not recommended. Remove such variables from dataset.
Continuous Variables:
Impute using Mean / Median
Categorical Variables:
New level can be created for Missing values. E.g. Unknown
Check for Co
elation:
Handle multicollinearity through variable transformation or derived variables.
Outlier Treatment:
Remove outlier as per domain understanding, impute using mean / median depending on variability.
Univariate Analysis Bivariate Analysis
Descriptive Statistics
EDA Insights
Dataset has 25000 rows / records and 24 Columns / variables
Data is collected between the age groups of 16 to 74 across Male & Female with occupation ranging from Student, Salaried and Business
Weight is ranging from 52kgs – 96kgs
“Alcohol intake” values ranging from No, Rare and Daily
“Doing Exercise” values ranging from Daily, Moderate and Extreme
Insurance Cost (Target Variable) is considered as Premium per Year; Insurance Cost is ranging from Rs 2468 to Rs 67870
Applicant_id column is i
elevant in the above context and hence can be ignored
Mean age = 44 and Max = 74
Mean BMI = 31 and Max = 100
16422 are Male (65%) and 8578 (35%) are Female
Models to consider for this project
Insurance Cost prediction is a linear regression problem( establishing a relationship between a dependent variable and one or more independent variables) , since we have to do continuous price prediction instead of logical operator type solution, so below is list of models that can be implemented.
Linear Regression – is one of the most common model for regression problems
Lasso The model is penalized for the sum of absolute values of the weights. Introduces a new hyperparameter, alpha, the coefficient to penalize weights.
Ridge It takes a step...
SOLUTION.PDF

Answer To This Question Is Available To Download

Related Questions & Answers

More Questions »

Submit New Assignment

Copy and Paste Your Assignment Here