MEDICAL
INSURANCE COST
PREDICTION
Using Machine Learning
Sahil Bangar
02713702021 M1
INTRODUCTION
• The project will analyze patient demographics, medical history,
and lifestyle factors to identify key drivers of insurance costs.
• Machine learning algorithms will be trained to predict insurance
costs based on the collected data, enabling personalized cost
estimates.
• The aim is to develop machine learning models to predict medical
insurance costs accurately. By leveraging various data sources
and advanced algorithms, our goal is to enhance decision-making
in healthcare financing and insurance management.
OBJECTIVE
• The objectives of this project are to develop accurate machine
learning models for predicting medical insurance costs.
• These models will be designed to enhance predictive accuracy,
ensure interpretability, and incorporate real-time predictions.
• The aim is to optimize model deployment, adhering to ethical
considerations throughout the project lifecycle, ultimately
improving decision-making in healthcare financing and
insurance management
ALGORITHMS USED
• Linear
Regression
• Random Forest
• XGBoost
ALGORITHMS USED
• Linear regression analysis is used to • Random Forest is a classifier that
predict the value of a variable based on contains a number of decision trees
the value of another variable. The
on various subsets of the given
variable you want to predict is called the
dependent variable. The variable you are
dataset and takes the average to
using to predict the other variable's value improve the predictive accuracy of
is called the independent variable. that dataset.
• Linear-regression models are relatively • Instead of relying on one decision
simple and provide an easy-to-interpret tree, the random forest takes the
mathematical formula that can generate prediction from each tree and based
predictions. Linear regression can be
applied to various areas in business and
on the majority votes of predictions,
academic study. and it predicts the final output.
ALGORITHMS USED
• XGBoost is an optimized distributed gradient boosting library
designed for efficient and scalable training of machine learning
models.
• It is an ensemble learning method that combines the predictions
of multiple weak models to produce a stronger prediction.
• One of the key features of XGBoost is its efficient handling of
missing values, which allows it to handle real-world data with
missing values without requiring significant pre-processing.
CHOICE AND WHY?
Linear Linear-regression models are relatively simple and provide
an easy-to-interpret mathematical formula that can generate
Regression
predictions. Linear regression can be applied to various areas
in business and academic study.
Random Forest
An ensemble method that combines multiple decision trees to
improve predictive accuracy, handle nonlinear relationships,
and capture complex interactions in rainfall data.
XGBoost
A powerful ensemble technique that enhances the predictive
performance of decision trees by using gradient boosting,
offering high accuracy, scalability, and feature importance
analysis for rainfall prediction models.
DATASET ANALYSIS
• Importing libraries & loading the dataset
• Information about data
TRAIN-TEST SPLIT
OUTPUTS
VISUALIZATIONS
THANK YOU!