🧠 Machine Learning Interview Q&A

✅ 1. What is Overfitting & Underfitting?
• Overfitting: Model performs well on training data but poorly on unseen data.
• Underfitting: Model fails to capture patterns in training data.
🔹 Solution: Cross-validation, regularization (L1/L2), pruning (in trees).

✅ 2. Difference: Supervised vs Unsupervised Learning?
• Supervised: Labeled data (e.g., Regression, Classification)
• Unsupervised: No labels (e.g., Clustering, Dimensionality Reduction)

✅ 3. What is Bias-Variance Tradeoff?
• Bias: Error due to overly simple assumptions (underfitting)
• Variance: Error due to sensitivity to small fluctuations (overfitting)
🎯 Goal: Find a balance between bias and variance.

✅ 4. Explain Confusion Matrix Metrics
• Accuracy: (TP + TN) / Total
• Precision: TP / (TP + FP)
• Recall: TP / (TP + FN)
• F1 Score: Harmonic mean of Precision & Recall

✅ 5. What is Cross-Validation?
• A technique to validate model performance on unseen data.
🔹 K-Fold CV is common: data split into K parts, trained/tested K times.

✅ 6. Key ML Algorithms to Know
• Linear Regression – Predict continuous values
• Logistic Regression – Binary classification
• Decision Trees – Rule-based splitting
• KNN – Based on distance
• SVM – Hyperplane separation
• Naive Bayes – Probabilistic classification
• Random Forest – Ensemble of decision trees
• K-Means – Clustering algorithm

✅ 7. What is Regularization?
• Adds penalty to model complexity
• L1 (Lasso) – Can shrink some coefficients to zero
• L2 (Ridge) – Shrinks all coefficients evenly

✅ 8. What is Feature Engineering?
• Creating new features to improve model performance
🔹 Includes: Binning, Encoding (One-Hot), Interaction terms, etc.

✅ 9. Evaluation Metrics for Regression
• MAE (Mean Absolute Error)
• MSE (Mean Squared Error)
• RMSE (Root Mean Squared Error)
• R² Score (Explained Variance)

✅ 10. How do you handle imbalanced datasets?
• Use techniques like:
• SMOTE (Synthetic Oversampling)
• Undersampling
• Class weights
• Precision-Recall Curve over Accuracy

👍 Tap ❤️ for more!