Obesity Prediction Model 🏥

A machine learning project that predicts obesity categories using lifestyle and physical condition data. The model classifies individuals into different obesity levels based on various behavioral and demographic features.

📊 Overview

This project implements a machine learning model to predict obesity levels based on eating habits, physical condition, and lifestyle choices. It uses a comprehensive dataset with both categorical and numerical features to make accurate predictions across seven different obesity categories.

✨ Features

• Multi-class obesity prediction
• Comprehensive data preprocessing
• Feature importance analysis
• Model performance evaluation
• Cross-validation implementation
• Handling of imbalanced classes
• Interactive visualizations

🔧 Requirements

• Python 3.9 or higher
• Required libraries:

  numpy>=1.21.0
  pandas>=1.3.0
  scikit-learn>=1.0.0
  seaborn>=0.11.0
  matplotlib>=3.4.0
  imblearn>=0.8.0
  

🚀 Installation

1. Clone the repository:

   git clone https://github.com/SnowAncestor/AI_Project.git
   cd AI_Project

2. Create and activate a virtual environment (recommended):

   python -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install required packages:

   pip install -r requirements.txt

💻 Usage

1. Later..

📝 Data Dictionary

Categorical Features

Feature	Description	Values	Encoding
Gender	Gender of the individual	Female, Male	0, 1
family_history_with_overweight	Family history of overweight	no, yes	0, 1
FAVC	Consumption of high-calorie food	no, yes	0, 1
CAEC	Eating between meals	Always, Frequently, Sometimes, no	0, 1, 2, 3
SMOKE	Smoking status	no, yes	0, 1
SCC	Calories monitoring	no, yes	0, 1
CALC	Alcohol consumption	Always, Frequently, Sometimes, no	0, 1, 2, 3
MTRANS	Transportation used	Automobile, Bike, Motorbike, Public_Transportation, Walking	0, 1, 2, 3, 4

Target Variable

Feature	Description	Categories	Encoding
NObeyesdad	Obesity level	Insufficient_Weight, Normal_Weight, Obesity_Type_I, Obesity_Type_II, Obesity_Type_III, Overweight_Level_I, Overweight_Level_II	0, 1, 2, 3, 4, 5, 6

Numerical Features

• Age (years)
• Height (meters)
• Weight (kilograms)
• FCVC (Frequency of vegetable consumption)
• NCP (Number of main meals)
• CH2O (Water consumption)
• FAF (Physical activity frequency)
• TUE (Time using technology devices)

📈 Model Performance

The current model achieves the following metrics:

• Accuracy: [Later]

🔄 Model Pipeline

1. Data Preprocessing

   • Handling missing values
   • Feature encoding
   • Feature scaling
   • Class balancing using SMOTE

2. Feature Engineering

   • BMI calculation
   • Feature interactions
   • Polynomial features

3. Model Training

   • Cross-validation
   • Hyperparameter tuning
   • Model evaluation

🙏 Acknowledgments

• Training data = 85%, Testing = 15%
• [Dataset Source](In The Data File)