Obesity Prediction – Classification

Objective

To predict whether a person is obese or not using health and lifestyle data, by training a machine learning classifier.

Dataset

Features include:

Age

Gender

Height and weight¹

Daily physical activity levels

Dietary habits (e.g., high-calorie food consumption)

Other lifestyle factors (e.g., smoking, alcohol)

The target is a binary label: obese vs not obese.

? Methods

Data cleaning and preprocessing (handling missing values, scaling)

Feature engineering (e.g., BMI calculation)

Exploratory data analysis (to understand patterns in features)

Training classifiers such as:

Logistic Regression

Decision Trees

Random Forests

K-Nearest Neighbors

Evaluating models using accuracy, precision, recall, and F1-score

Outcome

A trained model that predicts obesity status based on input features

Insights into which factors (e.g., physical activity, diet) play the biggest role

A tool that can be used for early detection and prevention recommendations