Must — know differences in Data Science
“Concepts are the keystone in the architecture of our thinking” — Paul Hughes
1. Bias vs Variance
2. Bagging vs Boosting
3. Random Forest vs Decision Trees vs Bagging
Random Forests add more randomness, as sampling is done at node level as compared to Bagging where a fixed set of features are a given base model (model level)
Hence, Bagging with underlying algorithm being decision trees is not the same as Random Forest.
4. Standardization vs Normalization
5. K-means vs Hierarchical Clustering vs K-means ++
6. K-means vs KNN
7. Parametric vs Non parametric
8. Precision vs Recall
Recall / Sensitivity : Of all the predicted positive cases by the model, how many were actually positive
Precision : Of all the positive cases, how many were correctly predicted
Precision- Recall trade-off
9. Mean-squared error vs Mean absolute error
10. R-square vs Adjusted R-square
11. Ridge vs Lasso Regularization
12. Parameters vs Hyperparameters
13. Batch vs Online Learning
14. Standard Deviation vs Inter Quartile Range
15. Correlation vs Causation vs Covariance
16. Probabilistic vs Non-Probabilistic Sampling
17. Overfitting vs Underfitting
18 . fit vs fit_transform
19. Type I vs Type II error
20. Supervised vs Unsupervised Leaning
21. Z-test vs T-test
Final Thoughts
Sometimes even simple terms and concepts get blurred with time while we keep focusing on more advanced stuff. It is thus crucial to have a compilation these basic foundational concepts while we continue to explore further
Hope this article helped!
Please feel free to add in your thoughts and suggestions (if any)
Happy learning!
Cheers :)