Water Potability Classification Using Machine Learning Algorithms

---

📌 Dataset Information

• Source: Kaggle Water Potability Dataset
• Features: pH, Hardness, Solids, Chloramines, Sulfates, Conductivity, Organic Carbon, Trihalomethanes, Turbidity, Potability
• Rows: 11,115
• Missing Values: pH, Sulfates, Trihalomethanes

---

Data Preprocessing

• Missing values imputed with mean strategy
• Outliers handled using IQR (Interquartile Range) method

---

Models Implemented

• Random Forest
• Decision Tree
• Logistic Regression
• K-Nearest Neighbors (KNN)
• Support Vector Machine (SVM)

---

---

🔥 Best Performing Model

✅ Random Forest with the highest accuracy and F1 Score.

---

📌 Conclusion

Random Forest emerged as the most effective model for predicting water potability, offering robust performance across all key evaluation metrics.

---