Data Classification in Data Science for Students

Data Classification in Data Science for Students

1. Classification

2. Models

3. Classification Target variable is categorical. Predictors could be of any data type. Algorithms Decision Trees Rule induction kNN Naive Bayesian Neural Networks Support Vector Machines Ensemble Meta Models

4. Decision Trees

5. Decision Trees Predictors / Attributes Target / Class

6. Decision Tree

7. Tree Split - Entropy

8. Measure of impurity Every split ties to make child node more pure. Gini impurity Information Gain (Entropy) Misclassification Error https://www.quora.com/What-are-the-advantages-of-different-Decision-Trees-Algorithms

9. Rule Induction

10. Tree to Rules Rule 1: if (Outlook = overcast) then yes Rule 2: if (Outlook = rain) and (Wind = false) then yes Rule 3: if (Outlook = rain) and (Wind = true) then no Rule 4: if (Outlook = sunny and (Humidity > 77.5) then no Rule 5: if (Outlook = sunny and (Humidity ≤ 77.5) then yes

11. Rules R = { r 1 ∩ r 2 ∩ r 3 ∩ .. r k } Where k is the number of disjuncts in a rule set. Individual disjuncts can be represented as r i = (antecedent or condition) then (consequent)

12. Approaches

13. Sequential covering

14. K Nearest Neighbors

15. Adage Birds of same feather flock together

16. Guess the species for A and B

17. KNN

18. Measure of Proximity Distance

19. Measure of Proximity Correlation similarity Simple matching coefficient Jaccard Similarity Cosine similarity

20. NAÏVE BAYESIAN

21. Predict your commute time http://www.wired.com/2015/08/pretty-maps-bay-area-hellish-commutes/#slide-2

22. Bayes’ theorem Class conditional probability Posterior probability Probability of the outcome Probability of the conditions

23. Data set

24. Class conditional probability

25. Test record

26. Calculation of posterior probability P( Y /X)

27. Issues Incomplete training set -> Use laplace correction Continuous numeric attributes -> Use Probability density function Attributes independence -> remove correlated attributes

28. NEURAL NETWORKS

29. Model Y = 1 + 2X1 + 3X2 + 4X3

30. Neurons

31. SUPPORT VECTOR MACHINES

32. Boundary

33. Margin

34. Transforming linearly non-separable data

35. Optimal hyperplane

36. Ensemble Learners

37. Ensemble model Wisdom of the Crowd Meta learners = sum of several base models Reduces the model generalization error

38. Ensemble models