📱 Student Social Media Addiction Analysis

A data science project exploring the impact of social media usage on student health, sleep, and academic performance using clustering and predictive modeling techniques.

📌 Project Summary

This notebook analyzes a dataset of 700+ students from multiple countries to understand how social media behavior correlates with academic outcomes and well-being indicators. Through exploratory analysis, clustering (K-Means), and logistic regression modeling, we uncover key behavioral patterns and risk profiles.

---

📁 Project Structure

• social-media-clustering-prediction-2.ipynb — Main notebook with EDA, clustering, and predictive modeling.
• README.md — Project overview and usage instructions.

---

🔍 Techniques Used

• Exploratory Data Analysis (EDA)
• K-Means Clustering
• Logistic Regression
• Correlation Heatmaps & ROC Curves
• Data Preprocessing (Label Encoding, Feature Scaling)

---

📊 Key Insights

• Strong correlation between high usage and poor sleep/mental health.
• K-Means revealed 3 distinct student usage clusters (light, moderate, heavy).
• Logistic regression showed significant predictors of academic impact (e.g., sleep hours, platform time).
• ROC-AUC of ~0.80 indicated solid predictive performance.

---

🛠️ Technologies

• Python (Pandas, NumPy, Scikit-learn, Matplotlib, Seaborn)
• Jupyter Notebook
• Git & GitHub

---

🚀 How to Run

1. Clone the repository

git clone https://github.com/your-username/social-media-addiction-analysis.git
cd social-media-addiction-analysis