Sleep-Health-and-Lifestyle Project

Overview

This project presents an analysis of a synthetic dataset containing sleep, cardiovascular metrics, and lifestyle factors for nearly 400 fictive individuals. The data explores various aspects such as sleep patterns, physical activity, stress levels, and their impact on sleep health.

Dataset

The dataset is provided in a CSV file, data.csv, containing columns like Person ID, Gender, Age, Occupation, and various health and lifestyle metrics.

Source

The dataset is sourced from Kaggle: Sleep Health and Lifestyle Dataset (https://www.kaggle.com/datasets/uom190346a/sleep-health-and-lifestyle-dataset/).

Scenarios

Identifying Potential Sleep Disorders

• Objective: Develop a classifier to predict sleep disorders based on lifestyle and health metrics.
• Background: For a health insurance company aiming to determine client premiums based on potential sleep disorders.

Exploratory Data Analysis (EDA)

The EDA includes visualizations and analyses like:

• Relationship between occupation and sleep patterns.
• Age, sleep duration, and sleep disorders.
• Sleep quality across different sleep disorders.
• Gender distribution in sleep disorders.
• Impact of physical activity on sleep health.

Feature Engineering

• Standardization and encoding of variables.
• Reduction of multicollinearity.
• Preparing the dataset for predictive modeling.

Modeling

Decision Tree Classifier

• Baseline model establishment.
• Hyperparameter tuning using Grid Search.
• Optimized Decision Tree Classifier.
• Confusion Matrix Analysis for performance interpretation.

Conclusion

The project concludes with insights into model performance and its application in medical decision-making.

Usage

• Clone the repository.
• Install required dependencies.
• Run Jupyter Notebooks to explore the dataset and models.