This project aims to predict insurace charges based on a set of features, such as age, gender, BMI and other factors. The goal is to create a machine learning model that can accurately estimate the charges for a given individual based on their profile.

The dataset used for this project is insurance.csv. It includes the following featues:

• age: Age of the individual
• sex: Gender
• BMI: Body Mass Index
• children: Number of children
• smoker: Whether the individual is a smoker
• region: The region where the individual resides
• charges: Medical insurance charges (target variable)

The preprocessing pipeline includes the following steps:

• Handling duplicated rows
• Handling missing values
• Correcting datatype
• Encoding categorical variables (sex, smoker, region)
• Feature scaling (StandardScaler for age, bmi, charges)

Linear Regression

• R-squared (R²)

• Model was validated with a new dataset called validation_dataset.csv

The mean R-squared scores across 5 folds is 0.75. This suggests a good level of predictive power, but may be further improved with hyperparameter tuning, feature engineering and exploring different model architectures.